Pest management of rice farmers in Asia - IRRI books - International ...

Pest Management
of Rice Farmers
in Asia
Edited by
K.L. Heong and
M.M. Escalada
1997
IRRI
INTERNATIONAL RICE RESEARCH INSTITUTE

The International Rice Research Institute (IRRI) was established in 1960
by the Ford and Rockefeller Foundations with the help and approval of the
Government of the Philippines. Today IRRI is one of 16 nonprofit intemational
research centers supported by the Consultative Group on International
Agricultural Research (CGIAR). The CGIAR is cosponsored by the
Food and Agriculture Organization of the United Nations (FAO), the International
Bank for Reconstruction and Development (World Bank). the United
Nations Development Programme (UNDP), and the United Nations Environment
Programme (UNEP). Its membership comprises donor countries,
international and regional organizations, and private foundations.
As listed in its most recent Corporate Report, IRRI receives support,
through the CGIAR, from a number of donors including UNDP, World
Bank, European Union, Asian Development Bank, and Rockefeller Foundation,
and the international aid agencies of the following governments:
Australia, Belgium, Canada, People’s Republic of China, Denmark, France,
Germany, India, Indonesia, Islamic Republic of Iran, Japan, Republic of
Korea, The Netherlands, Norway, Philippines, Spain, Sweden, Switzerland,
United Kingdom, and United States.
The responsibility for this publication rests with the International
Rice Research Institute.
The designations employed in the presentation of the material in
this publication do not imply the expression of any opinion whatsoever on
the part of IRRI concerning the legal status of any country, territory, city, or
area, or of its authorities, or the delimitation of its frontiers or boundaries.
Copyright International Rice Research Institute 1997
Los Baños, Philippines
Mailing address: P.O. Box 933, Manila 1099, Philippines
Phone: (63-2) 845-0563, 844-3351 to 53
Fax: (63-2) 891-1292, 845-0606
Email:
Postmaster@IRRI.CGNET.COM
Telex:
(ITT) 40890 Rice PM; (CWI) 14519 IRILB PS
Cable:
RICEFOUND MANILA
Home page: http://www.cgiar.org/irri
Riceweb: http://www.riceweb.org
Riceworld: http://www.riceworld.org
Courier address: Suite 1009, Pacific Bank Building
6776 Ayala Avenue, Makati
Metro Manila, Philippines
Tel. (63-2) 891-1236, 891-1174, 891-1258, 891-1303
Suggested citation:
Heong KL, Escalada MM, editors. 1997. Pest management of rice farmers
in Asia. Manila (Philippines): International Rice Research Institute. 245 p.
EDITORS: Bill Hardy, Carolyn Dedolph
EDITORIAL ASSISTANCE: Rogelio R. Quintos, Crisanta Culala
LAYOUT AND DESIGN: Ariel Paelmo
FIGURES AND ILLUSTRATIONS: Ariel Paelmo
COVER DESIGN: Juan Lazaro IV
ISBN 971-22-0102-3

Contents
FOREWORD
Kenneth S. Fischer
PREFACE
K.L. Heong, M.M. Escalada
ACKNOWLEDGMENTS
v
vii
ix
CHAPTER 1
Methods for research on farmers' knowledge, attitudes,
and practices in pest management
M. M. Escalada and K. L. Heong
CHAPTER 2
Pest management practices of lowland rice farmers in Cambodia
G.C. Jahn, P. Sophea, K. Bunnarith, and P. Chanthy
CHAPTER 3
Pest management practices of rice farmers in Hunan, China
Li Shae Sho, Guo Yu-Jie, Hu Guo-Wen, and Liang Di-Yun
CHAPTER 4
The role of women in rice pest management in Zhejiang, China
Hu Ruifa, Cheng Jiaan, Dong Shouzhen, and Sun Yinyin
CHAPTER 5
Pest management practices of rice farmers in Tamil Nadu, India
S.D. Sivakumar, S.R. Subramanian, S. Suresh, and M. Gopalan
CHAPTER 6
Pest management practices of rice farmers in West Java, Indonesia
S. Kartaatmadja, J. Soejitno, and I.P. Wardana
1
35
53
63
75
87
iii

CHAPTER 7
Pest management practices of rice farmers in the rainfed lowland
environment of the Lao PDR
H.R. Rapusas, J.M. Schiller, and V. Sengsoulivong
CHAPTER 8
Pest management practices of rice farmers in the Muda and
Kemubu irrigation schemes in peninsular Malaysia
R. Normiyah and P.M. Chang
CHAPTER 9
Farmers' perceptions of rice tungro disease in the Philippines
H. Warburton, EL. Palis, and S. Villareal
CHAPTER 10
Weed management practices of rice farmers in Iloilo, Philippines
K. Moody, M. M. Escalada, and K. L. Heong
CHAPTER 11
Pest management perceptions and practices of farmers growing rice
and vegetables in Nueva Ecija, Philippines
K. L. Heong, A.A. Lazaro, and G. W. Norton
CHAPTER 12
Pest management practices of rice farmers in Sri Lanka
L. Nugaliyadde, T. Hidaka, and M.P. Dhanapala
CHAPTER 13
Farmers' perceptions and practices in weed management in Thailand
L. Meenakanit and P. Vongsaroj
CHAPTER 14
The changing role of women in rice pest management in central Thailand
L. Meenakanit, M.M. Escalada, and K.L. Heong
CHAPTER 15
Pest management perceptions and practices of rice farmers
in Long An Province, Vietnam
Vo Mai, N. H. Huan, K. L. Heong, M.M. Escalada, and A.A. Lazaro
CHAPTER 16
A comparative analysis of pest management practices of rice farmers in Asia
K. L. Heong and M.M. Escalada
99
115
129
143
161
171
185
201
215
227
iv
Contents

Foreword
IRRI researchers have carried out intensive studies for many years in different rice
ecosystems on interactions among plants and their pests and predators. The economic
and health impacts of various control tactics have also been investigated. Based on
these findings, IRRI strongly endorses integrated pest management (IPM) based on
ecological principles in which natural mechanisms and processes, including hostplant
resistance, are fully exploited to reduce economic and yield losses to farmers.
IRRI works closely with national and international organizations to promote and
implement IPM by generating the knowledge needed, including an understanding of
farmers’ decision-making processes.
One of the goals of IPM is to eventually have a place in sustainable agricultural
systems with minimal or no synthetic inputs. IRRI agrees with this goal and believes
that farmers need to be given information to make decisions on pest management
techniques, and therefore supports policy changes that contribute to this empowerment.
In its continuing effort to reduce pesticide misuse, IRRI embarked on a series of
studies to understand farmers’ pest management knowledge, attitudes. and practices
in Asia. This research was designed and conducted through partnerships with national
agricultural scientists. The primary objective was to listen to farmers and understand
the various factors that constrain pest management decisions and practices
on-farm. Some critics argue that despite the flurry of research in pest management,
farmers’ decisions and practices in the past three decades have remained relatively
unchanged. In the past, limited attention was given to farmers’ practices. This collaborative
research, coordinated by IRRI, represents a unique integration of disciplines
to address the issue, involving entomologists, plant pathologists, weed scientists,
economists, and sociologists. Through a better understanding of how pest management
decisions are made and how new research information and practices can be
communicated, scientists can focus on developing more appropriate tools, methods,
and systems that are ecologically and economically sound for farmers to manage
diverse pest species. This book, which describes various methods used and reports the
v

findings of research carried out in 10 countries in Asia, will also serve as a baseline
for researchers to evaluate change.
IRRI is grateful to the Swiss Agency for Development and Cooperation (SDC)
for financial support for the Rice Integrated Pest Management Network, which coordinated
this research. The research was carried out by participants in national agricultural
research and extension systems, and we are grateful for their commitment to this
partnership.
KENNETH S. FISCHER
Deputy Director General for Research
vi Foreword

Preface
The theories and research underlying integrated pest management (IPM) in rice have
been well documented over the years in the scientific literature. Information about
current knowledge, attitudes, perceptions, and practices related to the pest management
of rice farmers in Asia, however, was relatively nonexistent. This book helps to
bridge that gap by providing the missing information from farmers’ perspectives.
In a sense, this volume is a report from farmers in the field, recorded through pest
management researchers. These professionals report on current practices of farmers
and the predicaments they face. The researchers also share farmers’ insights, experiences,
views, and recommendations for the future.
Experienced researchers and extension specialists from around Asia wrote the
chapters in this book. They carried out the surveys themselves, spending numerous
hours with farmers to observe their practices and discuss their pest-related problems.
We have targeted four specific audiences for this volume:
• Researchers and extension specialists in pest management or related fields
who direct or are involved in studies of farmers’ knowledge, attitudes, and
practices.
• Academics and trainers who prepare others to design and conduct farmer surveys
or to use the results.
• Extension workers who carry out pest management efforts, conduct farmer
training programs, and design communication materials.
• Agricultural administrators and policymakers who design and implement national
or regional policies related to pest management.
The chapters have been developed from a series of surveys and authors’ personal
interviews with farmers in 10 Asian countries: Cambodia, China, India, Indonesia,
Lao PDR, Malaysia, the Philippines, Sri Lanka, Thailand, and Vietnam.
The first chapter discusses the methods used in studying farmers’ knowledge,
attitudes, and practices, drawing on the lessons learned in conducting on-farm surveys,
focus group interviews, and personal interviews. The next two chapters examine
farmer pest management practices in Cambodia and Hunan Province in China.
vii

Chapter 4 features the role of Chinese women in pest management in Zhejiang Province.
Chapters 5, 6, and 7 describe general farmer practices in Tamil Nadu, India; West
Java, Indonesia; and the rainfed lowlands of Lao PDR. In Chapter 8, the Malaysian
authors compare pest management of farmers from two irrigation schemes, drawing
on similarities and differences. The next three chapters focus on the Philippines, featuring
farmers’ perceptions of the viral disease tungro, weed management practices in
Iloilo, and a comparison of farmer practices in both rice and vegetables in San Jose.
Chapter 12 discusses rice farmers’ pest management practices in Sri Lanka and
Chapter 13 features the weed management of farmers in Central Thailand. Chapter 14
discusses the changing role of women farmers as they begin to take on more farming
responsibilities. Chapter 15 presents details of farmers’ pest management perceptions
and practices in Long An Province, Vietnam. Finally, in Chapter 16, a comparative
analysis of practices, perceptions, and changes anticipated is given.
K.L. HEONG
M.M. ESCALADA
viii Preface

Acknowledgments
We are delighted to have had the opportunity to prepare this volume. Between 1990
and 1996, the Swiss Agency for Development and Cooperation (SDC) funded the
Rice IPM Network, which was coordinated by the International Rice Research Institute
(IRRI). Through this Network, a series of workshops was held during which
researchers developed and refined instruments to measure farmers’ knowledge, attitudes,
and practices. After pretesting these instruments, researchers implemented surveys,
the results of which are reported here. The Network provided partial funds for
all the surveys except for those conducted in Cambodia, Sri Lanka, and San Jose,
Nueva Ecija, Philippines.
Many people we have worked with over the years have contributed greatly to our
understanding of farmers’ pest management and deserve to be acknowledged. Among
them are Dr. Hans Peter Mag, Mr. Jurgen Benz, and Ms. Christine Grieder of SDC,
who provided their support for the Rice IPM Network; Dr. Klaus Lampe, former
IRRI director general, for his leadership in fostering an appropriate environment in
which to pursue the research; Dr. Ken Fischer, IRRI deputy director general for research,
for his encouragement and contributions in discussions; Dr. Dale Bottrell,
former head of the IRRI Entomology Division, and Dr. T.W. Mew, head of the IRRI
Entomology and Plant Pathology Division, for their leadership and encouragement;
and Dr. Sam Fujisaka, anthropologist, for teaching us how to talk and listen to farmers.
We are grateful to the contributors who dedicated their time to do the research,
analyze the data, and write the papers. Our thanks also go to the respective directors
general and heads of departments of the collaborating institutions for permitting their
staff members to contribute to this book, to Ms. Amor Lazaro, who assisted in data
management and reanalysis in many manuscripts, to Ms. Elena Genil, who typed many
of the manuscripts, and to the many enumerators who helped gather the data.
Last but not least, we express our sincere gratitude to the thousands of farmers
who responded to the survey questions.
ix

CHAPTER 1
Methods for research on farmers’
knowledge, attitudes, and practices
in pest management
M.M. Escalada and K.L. Heong
If researchers are to design appropriate improvements in pest management,
they must understand why farmers do the things they do.
The most common methods used to obtain this information include
rapid rural appraisals, key informant interviews, focus group interviews,
and farmer surveys. If properly planned and executed, these
methods can be cost-effective ways to obtain data for setting research
priorities and designing intervention strategies to improve
farmers’ pest management decision making.
The key informant interview involves talking to knowledgeable
persons to provide background information needed for designing a
survey. The focus group interview, a directed group discussion with
various sectors of a population, can be used to obtain insights into
farmers’ perceptions and concerns and to probe the issues that
emerged during a key informant interview. The farmer survey comprises
problem identification, questionnaire development, pretesting,
sampling, fieldwork, data processing, and analysis. It can be
used to improve problem definition, to raise further questions, and
to answer others.
Introduction
Research and development in pest management does not always lead to improved
practices. In the case of rice, farmers’ practices have changed little during the past
three decades, although research advancements have occurred (Brader 1979, Heong
and Sogawa 1994, Heong 1996). According to Norton and Mumford (1993), this
phenomenon may be attributed to two main problems. First, research and development
may be aimed at answering the wrong questions or at developing inappropriate
practices: a design problem. Second, research and development may be well targeted,
but the results are not getting through to be implemented by farmers: a delivery problem.
1

Pest organisms are sometimes fascinating subjects that can lead to research driven
by scientific curiosity — rather than the quest to solve problems. Although some basic
research can result in innovations, they must still be applied within farmers’ contexts.
Perhaps the low success rate of efforts to improve farmers’ pest management may be
because the problem of implementation is not seen as an integral part of the general
research and development agenda of designing appropriate improvements in pest
management (Norton and Heong 1988). The need — for both researchers and extension
workers — to understand why farmers do the things they do has been translated
into several research procedures (Byerlee et al 1980, Collinson 1984, Conway 1985,
Norton and Mumford 1993, Merrill-Sands and Collion 1993). These procedures emphasize
identifying the key questions by describing major components and relationships
involved in the problem.
Farmer surveys are used to improve problem definition, to raise further questions,
and to answer others. Other techniques have been used to complement farmer
surveys, such as the rapid rural appraisal and historical and seasonal information, as
have other participatory approaches, such as problem specification workshops. During
these workshops, stakeholders exchange knowledge and experience to determine
pest problems, identify areas to work on, and develop action plans for addressing the
problems (Bilston et al 1997).
This chapter describes farmer surveys and the various techniques that researchers
are using to obtain information about farmers’ knowledge, attitudes, and practices,
and the reasons for not implementing new practices endorsed by research results.
A farmer survey is an important data-gathering process for assessing the needs of
intended beneficiaries to determine their knowledge and perceptions of a pest problem,
their constraints in dealing with the problem, and their attitudes and practices in
pest management. The role of such surveys is to determine the nature of farmer practices
that can guide both research and extension work. If carefully designed and implemented,
farmer surveys can identify gaps in knowledge, misconceptions, or inappropriate
practices that need to be addressed by research (Bentley and Andrews 1996).
The scope of a farmer survey need not be limited to farmers' pest management knowledge,
attitudes, and practices, but it could also cover aspects such as decision-making
patterns, agronomic practices, and socioeconomic profiles.
Farmer surveys, also referred to as baseline surveys or knowledge-attitude-practice
surveys, have been done in recent years in Food and Agriculture Organization of
the United Nations (FAO)-supported strategic extension campaigns in Asia, Africa,
and the Caribbean. Topics covered weed management, pest surveillance, rat control,
golden snail management, temperate fruit crop cultivation, and maize production
(Adhikarya 1994). The data derived from such surveys have been used to define problems,
plan programs, design strategic extension messages, and serve as benchmark
data for a subsequent summative evaluation. Besides its use in extension planning,
the knowledge-attitude-practice survey can serve as a diagnostic tool to identify the
real problem and decide whether an extension campaign is a suitable solution. A
2 Escalada and Heong

Fig. 1. Using a survey on knowledge, attitudes, and practices to explore
solutions in an extension approach.
series of exploratory questions to help in making decisions using knowledge-attitudepractice
surveys is provided (Fig. 1).
Methods to solicit farmer information
Surveys tend to generate quantitative information that is only surface-deep; they should
therefore be complemented with more sensitive and in-depth methods (Chambers
1983, Bentley and Andrews 1996). Informal qualitative methods, such as diagnostic
surveys, focus group interviews, and key informant interviews, have been employed
in some of the more innovative and cost-effective farmer surveys. In most cases, a
diagnostic survey and key informant interviews often precede a formal baseline survey
(Fujisaka 1991). These activities help structure the subsequent formal survey and
ensure that it is focused and appropriate in the local context.
Table 1 outlines the main features of data collection methods (key informant
interview, focus group interview, and formal survey) used to complement a farmer
survey according to objectives, sampling procedure, type of instrument, data collection
method, and data processing and analysis.
Issues that may emerge during a key informant interview can be probed in the
focus group interview. While the key informant interview can provide leads, the focus
group interview can be used to clarify points raised and explore whether there is a
consensus on the concerns voiced by key informants. The research areas can be probed
in focus groups to help generate ideas and develop hypotheses that will then be fully
assessed in the larger, quantitative study. For example, in planning a major national
Methods for research on farmers' knowledge, attitudes, and practices in pest management 3

Table 1. Differences among data collection methods.
Steps
Key informant
interview
Focus group
interview
Formal survey
Objective Explore related Get a consensus Determine levels
insights
of knowledge,
attitudes, and
practices
Sampling Purposive Purposive Random
procedure
Instrument Guide questions Interview guide Structured
questionnaire
Data Unstructured Group discussion Structured
collection personal interview personal
interview
Data Mainly qualitative Mainly qualitative Mainly
processing
quantitative
and analysis
survey on farmers’ knowledge, attitudes, and practices related to weed management,
separate focus groups might be conducted with farmers to identify key issues. Focus
groups could help researchers generate hypotheses and develop the wording for specific
questions to be used in a formal survey.
Key informant interviews
Used as a tool to explore related issues and problems associated with a given topic, a
key informant interview involves talking to persons such as extension workers, key
farmers, local government officials, traders, and community leaders who know the
area or certain aspects of the problem (Jimenez 1985). From these key informants,
researchers can obtain information about the most important production problems,
cropping patterns, rainfall distribution, and other relevant topics that can help them
make decisions on subsequent activities. Taking an unstructured interview approach,
this type of interview enables the researcher to gain new insights, raise questions, and
examine phenomena from different perspectives (Bogdan and Taylor 1975, Okamura
1985). These informal interviews are useful for providing background information
for defining the issues to be addressed by a formal survey and as a guide for developing
a more structured questionnaire (Bryman 1988, Siebert 1973). An interview guide
that contains questions to gather information relevant to a certain issue may be used
to carry out the key informant interview.
4 Escalada and Heong

Focus group interviews
The focus group interview is a qualitative research technique originally developed to
give marketing researchers a better understanding of the data from quantitative consumer
surveys. As an indispensable tool for marketing researchers (Krueger 1988),
the focus group interview has become extremely popular because it provides a fast
way to learn from the target audience (Debus 1988, US Department of Health and
Human Services 1980). This technique is a rapid assessment, semistructured datagathering
method in which a purposively selected set of participants gathers to discuss
issues and concerns based on a list of key themes drawn up by the researcher or
facilitator (Kumar 1987). Marketing and media studies have shown that the focus
group interview is a cost-effective technique for eliciting views and opinions of prospective
clients, customers, and end-users. In agriculture, these interviews have been
used to obtain insights into target audience perceptions, needs, problems, beliefs, and
reasons for certain practices.
The focus group interview guide. To keep the session on track while allowing
respondents to talk freely and spontaneously, the facilitator uses an interview guide
that lists the main topics or themes to be covered in the session. It serves as a road
map that guides the facilitator in covering the list of topics and in keeping the discussion
on track. The number of items in the guide is generally kept to a minimum to
leave enough time for in-depth discussion. It should focus only on relevant research
issues. The sequence of topics in the guide usually moves from general to specific.
The following steps are suggested for developing the focus interview guide:
1. Specify the objectives and information needs of the focus group interview.
Example
To determine farmers’ reasons for following certain weed management practices.
2. Break down the major topics into discussion points or themes.
Example
a) Reasons for using weed management options
• chemical
• cultural
• mechanical
• biological
b) Reasons for nonadoption of nonchemical weed management options
c) Reasons for applying water management at particular crop stages to control
weeds
3. Prepare probe questions.
Example
a) Let’s talk about the reasons farmers use certain weed management practices
in this district. Why do farmers prefer to use herbicides instead of
nonchemical options?
b) Do you know of farmers in your area who do not practice weed control? In
your opinion, what are the reasons these farmers do not like to control
weeds in their ricefield?
Methods for research on farmers’ knowledge, attitudes, and practices in pest management 5

4. Review the guide and eliminate any irrelevant questions.
Asking questions during focus groups. The quality of questions asked in a focus
group can make a large difference in the kind of information obtained. Krueger (1988)
gives some tips on how to handle open-ended and dichotomous questions in these
interviews.
Open-ended questions are most appropriate at the start of the discussion because
they allow participants to answer from different angles. As the possible responses are
not preconceived, open-ended questions give the participants opportunities to express
their thoughts and feelings based on their specific situations. Krueger warns that some
questions may appear to be open-ended but are really closed-ended because they include
phrases such as “satisfied,” “to what extent,” and “how much.”
Dichotomous questions are ones that can be answered with a “yes” or “no” or
other similar two-alternative choice. As yes-no questions are dead ends, they usually
do not trigger the desired group discussion. They also tend to elicit vague responses
that do not lead to an understanding of the key issues being discussed (Moulton and
Roberts 1993).
Logistical arrangements
Invitations. Participants are contacted in advance, at least one to two weeks before the
session. A letter of invitation may be sent to each participant, depending on the prevailing
practices in the area. Participants are reminded about the focus group interview
one day before the session.
Group composition. The choice of participants depends on the focus group topic.
Often, the people who are included are those knowledgeable about the topic, but it is
also wise to gather the views of certain groups within the target population. The optimal
number of participants is 8-10, If a group is too small, one person in the group
may dominate it; if it is too big, then it may be difficult to control. Group members
should be representative of the target population.
Transportation. To ensure attendance, transportation is usually arranged for the
participants from their residences to the interview venue and back. In rural areas where
farm families may reside in distant villages, participants could be asked to converge
at a central location to facilitate pick-up.
Venue. Focus group interviews can be conducted in a place where 8-10 persons
can be seated and assured of some privacy. In rural areas, the most readily available
sites are school buildings, health and community centers, and churches. An appropriate
venue is a neutral place that is free from distractions and where participants can
talk openly.
Seating arrangement. 4 semicircular seating arrangement facilitates interaction
among participants because it allows them to freely see and hear each other.
Timing. The timing of the meeting should be convenient to all participants. While
waiting for others to arrive, the focus group interview team can use the time to break
the ice by getting information about the participants’ backgrounds. To minimize boredom,
focus group interviews are generally not stretched beyond two hours.
6 Escalada and Heong

Name tags. It is best to remember the names of the participants. Often, a seating
arrangement will facilitate identifying each one. If the culture permits, providing name
tags to participants is useful because it enables facilitators to call on those who may
be too shy to express their opinions.
Recording of discussion. A trained rapporteur should be asked to capture the
discussion in writing and note the participants’ nonverbal expressions. Situations may
occur where the discussion needs to be tape-recorded, but facilitators should weigh
the advantages and disadvantages.
Refreshments. When resources permit, serving refreshments after the session is a
small gesture of appreciation to the participants for having taken time off from their
work to participate.
How to conduct a focus group interview
Facilitator: In selecting a person to moderate a focus group, it is important that this
person have these qualities:
• familiarity with the discussion topic
• ability to speak the language of the area
• cultural sensitivity, including not acting as a judge or a teacher, not looking
down on respondents, not agreeing or disagreeing with what is said, and not
putting words in the participants’ mouths
• sensitivity to women
• politeness
Steps in conducting the session. Before the focus group interview begins, the
facilitator should obtain background information on the participants, such as age,
crops grown, and other pertinent items. The type of information to collect depends on
the topic. Once this is done, the following sequence of steps is carried out:
1. After a brief introduction, the purpose and scope of the discussion are explained.
2. Participants are asked to give their names and brief background information
about themselves.
3. The discussion is structured around the key themes using the probe questions
prepared in advance.
4. During the discussion, all participants are given the opportunity to participate.
Data analysis
After conducting the focus group interview, the key findings are described, analyzed.
and written up in a report. See Appendix A for a sample content page. Debus (1988)
suggests some useful guidelines for analyzing data:
1. Develop a plan for analysis consisting of
• background of the research
• objectives
• methods
• discussion details
• focus group interview guide
Methods for research on farmers’ knowledge, attitudes. and practices in pest management 7

2. Analyze the content of the group discussion by
• reviewing the notes from the focus group
• listening again to the cassettes from the session (if tape-recorded)
• grouping research findings according to key themes
• identifying the different positions that emerged under each key theme
• summarizing each of the different positions and assessing the extent to
which each position was held by participants
• pulling out verbatim phrases that represent each position
3. Synthesize the group discussion by
• reviewing the moderator’s notes of each discussion
• identifying the recurrent ideas that came out during each discussion
• interpreting these recurrent ideas based on other findings that emerged
• identifying the differences expressed for each topic and summarizing the
findings and group discussion.
Farmer surveys
In a survey, a researcher collects data from a sample of a population to determine the
relative incidence, distribution, and interrelations of variables for the purpose of describing
or predicting and as a guide to action (Kerlinger 1986, Oppenheim 1966).
Sample surveys focus on people and their beliefs, opinions, attitudes, motivations,
and behaviors. The goal of survey research is to infer the characteristics of a population
from samples drawn from that population.
Limitations of the survey approach
Chambers (1983) summarizes some of the problems with the survey approach:
“...questionnaire surveys often take more time and resources than estimated, enslave
researchers, and generate misleading data and unread reports. Some bad questionnaire
surveys make rural people appear ignorant when they are not.”
Because of the usefulness of surveys in setting the research agenda, designing
extension strategies, and evaluating the effectiveness of projects and development
interventions, researchers have often used them to get the information they need.
Surveys are also popular because their data can be subjected to a statistical analysis,
which generates quantitative indicators that give the semblance of rigor often desired
by donors and practitioners. But surveys can be costly, inefficient, and superficial
(Bryman 1988, Chambers 1983, Kearl 1976) unless they are carefully planned and
combined with in-depth, more sensitive techniques used by a multidisciplinary team
(Yin 1984, Bryman 1988, Gonzalez 1985).
How to conduct a farmer survey
It is important that surveys be planned and conducted with the utmost care because
their cost-effectiveness and opportunity costs are often high (Chambers 1983). The
following procedure is recommended:
8 Escalada and Heong

Table 2. Farmers’ sources of rice seeds, last cropping season,
central Thailand, 1996.
Source
Farmers (no.) (%)
Private seed grower
Neighbor
Own stock
Ministry of Agriculture
Identifying the problem. The first step in planning a farmer survey is to identify
the problem that needs to be addressed. In pest management, the choice of the specific
pest problem on which to focus depends on the research priorities and information
needs of a given ministry or plant protection organization. Where these priorities
have not been articulated, the farmer survey could also be used to gather information
for developing them.
Developing survey objectives. Once a priority problem has been identified, the
next step is to develop the survey objectives. A list of the variables that will help find
answers to the survey objectives may put the researcher on track. Similarly, specific
questions aimed at various aspects of the problem could help clarify the research
problem. It is also important to remember that the survey objectives should guide the
choice of questions.
Dummy tables for analyzing the data could help clarify the research problem and
guide the construction of interview questions. A sample dummy table on farmers’
sources of rice seeds is provided (Table 2).
Developing the survey instrument. In a farmer survey, the instrument used for
data collection is a questionnaire that contains a series of questions designed to gather
information from the respondents. Depending on the survey objectives, the questions
may focus on topics such as farmers’ knowledge, attitudes, and practices for a pest
problem, their cropping patterns, and their demographic and socioeconomic backgrounds.
Two types of questionnaire items are commonly used: closed-ended and openended.
Scale items are a third type.
Closed-ended items. Fixed-alternative or closed-ended items offer the respondent
a choice among two or more alternatives. For most items, closed-ended questions
are valuable because they provide a frame of reference and help to clarify the
meaning of the question. The most common type of closed-ended question is dichotomous,
which uses two-alternative answers, such as yes-no, agree-disagree. Often a
third alternative, don’t know or undecided, is added.
An example of one type of closed-ended question is
Who is responsible for water management in your ricefield?
1) self
2) spouse
Methods for research on farmers’ knowledge. attitudes. and practices in pest management 9

3) hired labor
4) other (specify)
With closed-ended items, a researcher can achieve greater uniformity and reliability
of measurement by encouraging respondents to answer in a way that fits preset
response categories—which facilitates data processing. But their major disadvantage
is their superficiality. Without probing questions, they do not ordinarily get beneath
the response surface. Probing questions are follow-up items that encourage respondents
to clarify their answers, giving more details about thoughts, feelings, and opinions
(Moulton and Roberts 1993). If they are mixed with open-ended items and used
with probing questions, closed-ended items are useful (Bryman 1988, Kerlinger 1986).
Open-ended items. The second type of item, open-ended questions, supplies a
frame of reference for respondents’ answers, but puts a minimum of restraint on answers
and their expression. Although their content is dictated by the research problem,
they impose no other restrictions on the manner in which respondents answer.
An example is
How should water management be carried out to control weeds?
Open-ended questions are flexible. They provide opportunities for getting indepth
information that enable the interviewer to clarify the question, ascertain a
respondent’s lack of knowledge, detect ambiguity, encourage cooperation and achieve
rapport, or make better estimates of respondents’ attitudes and beliefs (Kerlinger 1986).
Scale items. A scale is a set of verbal items to which an individual responds by
expressing degrees of agreement or disagreement or some other mode of response
(Kerlinger 1986, Kidder 1981, Becker 1970). Scale items have fixed alternatives and
place the responding individual at some point on the scale. Scales often consist of
response categories of multiple points, such as 7, 5, or 3. While a 5-point or 7-point
scale is generally suggested to minimize a response bias, a 3-point scale has been
observed to be more appropriate for rural settings because of differences in vocabulary
patterns and education levels. Rural respondents are often unable to put gradations
in their own perceptions or feelings, making it difficult to discriminate between
“agree” and “strongly agree.” An example of a scale follows:
Leaf-feeding insects do not cause yield loss.
1) strongly agree
2) agree
3) neither agree nor disagree
4) disagree
5) strongly disagree
6) don’t know
10 Escalada and Heong

For agree-disagree scales such as the one shown, it is important to include a
“don’t know” category so that respondents who have no knowledge of an issue can
say so. When using a “don’t know” category, we suggest that the midpoint of the scale
become “neither agree nor disagree.”
Assessing farmers’ knowledge, attitudes, and practices. In farmer surveys, some
questions seek to establish which particular technique or concept respondents know
about, how they feel about it, and whether they practice it. An objective norm is provided
against which to compare respondents’ knowledge and practices, and an objective
standard is given against which to rate the extent of their knowledge and practices,
and the direction of their attitudes. The assumption behind measuring the knowledge,
attitudes, and practices of respondents is that differences exist between what
people know (knowledge), how they feel (attitude), and what they do (practice).
Using the information gleaned, a researcher can identify gaps in knowledge, attitudes,
and practices for a specific problem, which may have implications for designing
future research and for implementing interventions. In a survey of farmers’ weed
management practices, for example, we found that most farmers relied on their own
seed or that of a neighbor for transplanting or broadcasting in their fields. Most believed
that seeds from private seed growers do not require additional cleaning, and
that neighbors and friends should exchange seeds among themselves. The farmers
said they preferred their own or their neighbors’ unprocessed seeds because a supply
is always available and accessible, and the cost is low. Information like this about
farmers’ knowledge can be transformed into researchable problems, with the research
results eventually communicated back to the farmers.
Appendix B provides a sample survey questionnaire containing questions on
knowledge, attitudes, and practices.
Knowledge, attitude, and practice questions defined. Knowledge questions are
constructed to test a respondent’s comprehension and awareness of the subject matter
of the survey.
Attitude questions measure how respondents feel about a particular subject. These
questions use scales to ask respondents to indicate their degree of agreement or disagreement
with certain statements. An example of an attitude scale is the following:
Direct seeding or broadcasting can increase the weed population.
_____1) agree
_____2) no opinion
_____3) disagree
Questions on practices measure the action component of a survey by asking respondents
how they behave or would behave in a particular situation. Although observation
is a more reliable method for determining a practice, farmer surveys often
measure a practice through verbal reports of a respondent’s use of a particular technique
or a set of recommended procedures. In asking about a respondent’s present
behavior or practice, specific—rather than general—questions are asked about what
the person is actually doing. In a survey on farmers’ pesticide use, for example, it is
Methods for research on farmers’ knowledge, attitudes, and practices in pest management 11

preferable to ask, “Which pesticide do you currently use? May I see the container?’
rather than merely asking, “Which chemicals do you use?”
Question wording. Often the validity and reliability of survey data depend on the
quality of the wording of the questions. Appropriateness and preciseness of the wording
are desirable attributes for generating reliable answers. To elicit accurate responses,
sensitivity must be given to cultural differences in linguistic patterns. For instance, in
pretesting of the survey instrument to determine message reach and use in Vietnam,
the following survey questions were found to be difficult to interpret:
26b. IF NO, why did you decide not to do the experiment?
27. Have you stopped early spraying for leaffolder control?
1) Yes
2) no
In this case, respondents perceived the use of negative words in a question to
imply a negative meaning (stopped the early spray for leaffolder control), so they
were considered inappropriate and culturally unacceptable. The questions were modified
to:
26b. Did you carry out the experiment?
If not, why not?
27. Did you spray insecticides in the early season for leaffolder
control?
When researchers work in cross-cultural survey settings, it is essential to exercise
care in the choice of words and syntax of survey questions. It is thus important to
pretest all questions before using them.
Being aware of the common errors in constructing questionnaires can help prevent
problems that threaten the validity of data:
1. Leading questions that suggest answers or force respondents to answer in the
direction desired by the researcher
Example
Do you think using IPM is a waste of time?
1) yes
2) no
An improvement: What do you think of IPM?
12 Escalada and Heong

2. Double-barreled items that ask two questions simultaneously
Example
What major pests did you have last season, and how did you control them?
An improvement:
a) What major pests did you have last season?
b) For each of the above, how did you control it?
3. Vague references to time
Example
Usually, what is your most important rice pest?
An improvement:
Last season, which pest did you consider to be the most important?
4. Subjective qualifiers
Example
Normally, how many times do you spray chemicals on your rice crop?
An improvement:
Last season, how many times did you spray chemicals on your rice crop?
5. Very long questions that tax the memory of a respondent
Example
For your last rice crop, could you tell me the dosage or number of tablespoons
per load, number of sprayer loads per hectare, time of application, and number
of applications of insecticides?
An improvement: Break this up into several questions:
Methods for research on farmers' knowledge, attitudes, and practices in pest management 13

In your last rice crop,
a) how many chemical applications did you use?
b) how many of these were insecticides?
c) how many sprayer loads per hectare did you use?
d) what was the dosage or number of tablespoons of insecticides used in each
load?
6. Jargon that confuses respondents
Example:
Which of the following is the best way to conserve natural enemies ?
_____1) spray on a fixed calendar schedule
_____2) spray at the first appearance of the pest
_____3) follow the economic threshold level method
An improvement:
Which of the following is the best way to conserve spiders (or use words farmers
normally use)?
_____1) spray on a calendar schedule
_____2) spray when the pest first appears
_____3) spray when the pest reaches a certain number.
Pretesting the questionnaire. After a researcher completes the draft of the prototype
questionnaire, it is pretested on a small representative sample of the population.
The pretest is a screening of the questionnaire to see how it works and whether changes
are necessary before the actual survey is started. About 15 to 25 respondents are adequate
for a pretest. Based on the pretest results, the questionnaire is modified and
then finalized.
Rationale for pretesting. The pretest provides a means of catching and solving
unforeseen problems in using the questionnaire, such as in the phrasing and sequencing
of questions. Linguistic and cultural differences often complicate the task of developing
a questionnaire—making pretesting indispensable.
The pretest enables a researcher to
• improve the wording of the questionnaire;
• correct and improve the translation of technical terms;
• check the accuracy and adequacy of the questionnaire's instructions, such as
skip and go to;
• eliminate unnecessary questions and add necessary ones; and
• estimate the time needed to conduct the interview.
Excerpts of pretest results are given in the box on the following page to illustrate
how pretesting yielded constructive suggestions that served as the basis for improving
a questionnaire.
14 Escalada and Heong

Questionare pretesting: suggestions from the field
The survey questionnaire for the project “Farmers’ perceptions and attitudes toward
seed health for crop management” was pretested in Pandac, Pavia, Iloilo, Philippines,
on 4 March 1995. The goals of the pretest were to determine the reactions of rice
farmers to the questionnaire, estimate the time needed to complete the interview,
validate the translation of key technical terms used, find out whether the respondents
could understand the technical terms, and ascertain whether the sequence of the
questions solicited the desired information. A survey enumerator was hired to assist
in pretesting the instrument with 15 rice farmers. The following observations and
corresponding recommendations were made:
1. In this village, the first and second rice crops are both grown during the wet season,
so the reference to the first one as the wet season crop and the second one
as the dry season crop does not apply. Recommendation: Provide the necessary
indicators for the interviewers to guide them if they are confronted with decision
problems in this area.
2. The second cropping was not realized because of drought (question 4: Last year
for the second cropping, what was the area on which you direct-seeded and/or
transplanted?). Some farmers did not plant rice and instead planted watermelon
or tomatoes. Recommendation: Point this out to the interviewers during the orientation.
3. In question 5 (What cropping pattern did you practice?), the farmers found the
term “cropping pattern” difficult to comprehend. The question cannot be properly
answered without explaining the term or making it a leading question. Recommendation:
Modify the question to instead ask what crop was grown in a particular
season. The farmers also mentioned a third cropping season during which they
plant crops such as watermelon, tomatoes, and mungbean in lieu of rice.
4. Question 6 (Last cropping season, how much (in pesos) did you spend for rice
seeds, insecticides, fungicides, herbicides, fertilizer, irrigation, and labor for pesticide
application?) entailed a lot of time to answer because the interviewer often
had to add up the figures given by farmers. Recommendation: Remind interviewers
to bring calculators with them.
Sampling techniques for choosing respondents. Deciding how many respondents
to include is an important concern in survey research. Standard social science methods
are used in selecting the sample for a farmer survey: cluster/multistage sampling,
stratified sampling, systematic sampling, and simple random sampling. The choice of
sampling technique depends primarily on the nature of the problem, the cost and time
factors involved, and the desired precision or reliability of the results (Parel et al
1978, Casley and Lury 1982).
Decisions on sample size depend on the degree of accuracy required, the degree
of variability in the population, and the kind of data analysis being planned. Sample
size can be estimated statistically or by following certain rules. The statistical method
requires some assumptions about the population and uses statistical equations for
random sampling processes. These procedures can be found in books on social science
research methods (Neuman 1997, Frankfort-Nachmias and Nachmias 1996).
Methods for research on farmers’ knowledge, attitudes, and practices in pest management 15

Rules for determining sample size are based on past experience with samples
drawn from statistical methods, and they are commonly used when studying large
populations. According to Neuman (1997), these rules are:
• A small population needs a larger sampling ratio to have an accurate sample.
For instance, a population of below 1,000 requires a sampling ratio of 30% to
achieve a high level of accuracy. For large populations (above 150,000), smaller
sampling ratios (1%) are often sufficient. In general, practical considerations,
such as costs and logistics, are the most important deciding factors.
• For small samples, small increases in sample size can produce larger gains in
accuracy than for large samples.
The following are descriptions of some of the commonly used sampling methods:
Cluster/multistage sampling. Most knowledge, attitude, and practice surveys use
the cluster or multistage sampling procedure to select respondents. This technique is
recommended when a complete list of persons in the population is not available and if
the population covers a large area. In this technique, the researcher comes up with the
desired sample by first selecting large groupings (clusters) within the population (Kidder
1981). The sampling is done successively in stages. Through simple or stratified
sampling, one chooses the clusters and then the units within each cluster. For example,
a survey of farming households may take a sample of regions; within each
region, a sample of provinces; within each province selected, a sample of districts;
and within each district selected, a sample of households (Parel et al 1966, Kidder
1981).
Stratified sampling. Sometimes, the population under study is large and made up
of diverse groups. Within the subset of rice farmers. further groupings exist, such as
irrigated, rainfed upland, and rainfed lowland. In such cases, stratified sampling is
suggested. When the composition of each group is not equal, the number of farmers
selected to be respondents should be proportional to the total number of farmers in
each of the three strata.
Systematic sampling. If the population units are within a limited area, a sample of
respondents may be selected systematically (Casley and Lury 1982). In systematic
sampling, the persons in the population from which the sample will be drawn are
numbered consecutively on a list. Then we determine the sampling interval by dividing
the sampling population by the desired sample size (N/n). We select the first sample
unit at random and choose the succeeding units according to the sampling interval.
Simple random sampling. Simple random sampling is done if there is a list of
persons to be included in the survey, if these persons are not widely spread out geographically,
and if they have similar characteristics. For instance, if a researcher is
doing a survey of fruit growers in the northern highlands of a province, he or she
needs a list of fruit tree growers in those communities. To select respondents through
simple random sampling, various techniques are suggested, such as drawing lots or
using random numbers (Parel et al 1966, Kidder 1981).
Implementing the field survey. Once the questionnaire has been pretested, finalized,
and reproduced, the next step is to implement the field survey. Resources needed
16 Escalada and Heong

for the fieldwork include personnel, money, and time. A survey coordinator, a supervisor,
and interviewers often make up a field survey team. The survey coordinator is
responsible for all aspects of the fieldwork, including selecting, training, and deploying
interviewers. The supervisor assists the survey coordinator in spot-checking and
monitoring the field interviews. Before they begin conducting the survey, interviewers
are oriented on the purpose of the survey and trained on interviewing skills and on
how to conduct the interviews. Guided by the sampling plan and respondent list, the
interviewers locate the respondents, conduct the interviews, and check the completed
questionnaires after the interview.
Choosing a field interviewer: Interviewers are important links in the survey chain,
so those who are selected must be honest and objective. We have found that college
students tend to be more objective interviewers because they do not have the inherent
bias that professional agency staff members often have. In a survey of rice farmers’
pest management perceptions and practices, for example, we observed that the plant
protection officers who had done the interviews tended to unnecessarily interpret farmers’
responses. Although many farmers reported that “green worm” was their most
important pest, interviewers recorded the pest as either armyworm or rice bug based
on their perception of what “green worm” implied. Green worm could have referred
to a variety of leaf feeders, such as rice leaffolders, cutworms, caseworms, and thrips.
Guidelines for interviewers when implementing the field survey. To ensure efficient
implementation of the survey and to minimize errors, the following guidelines
for interviewers are suggested:
Selecting respondents. Only those farmers who are on the respondent or replacement
list should be interviewed. If the designated farmer is temporarily not available
at the time of the interview, schedule a return visit. If the person will not be available
for a long time, choose a name from the list of replacements.
Materials. Advise interviewers to obtain the following before fieldwork: specific
area of assignment, list of rice farmers to be interviewed, questionnaires, a map, and
pencils.
Establish rapport with the respondent. As a quick approach for obtaining information,
a survey often relies on interviewers who are unknown to the respondents.
Being a stranger can pose difficulties—respondents often hesitate to give accurate
information because of an implicit mistrust of outsiders. When conducting field interviews,
a first step would be to establish rapport with respondents by setting the proper
atmosphere. To achieve this, the following steps are suggested:
1. Introduce yourself.
2. Ask the respondent’s permission for the interview.
3. Inform the respondent of these important points:
• description of the study
• purpose
• benefits that can be derived from the study
4. The following is an example of an introduction:
“Good morning, I am (insert name), a student at Long An University. The Plant
Protection Department (PPD) is currently doing a survey of farmers in Long An Prov-
Methods for research on farmers’ knowledge, attitudes. and practices in pest management 17

ince with regard to the recent multimedia campaign on pest management. We would
like to find out what you heard from the campaign and whether you told others about
it. Your feedback will help PPD and the People’s Committee make plans on how to
better respond to your information needs on pest management. May we please interview
you?”
How to conduct the field interview
1. Create a friendly atmosphere by being courteous, conversational, and unbiased.
When a respondent gives an answer that indicates a knowledge gap,
never show surprise or disapproval.
2. Ask each question exactly as it is worded in the questionnaire. If the respondent
has difficulty in understanding the question, however, either repeat the
question or try to adjust it slightly to facilitate understanding.
3. Ask questions in the order they appear on the questionnaire. Jumping from
question 1 to question 6 then back to question 2 will confuse the respondent
and affect the answers given.
4. Ask every question, unless the directions on the questionnaire specifically
state to skip certain ones. It is important that the questionnaire be filled out
completely. Record responses as they are reported to you. Do not give a personal
interpretation of a respondent’s answers.
5. Be extremely careful not to suggest a possible reply. When the respondent
does not give a quick answer, be patient and wait for the response.
6. Do not read aloud the response categories unless instructed to because they
will lead the respondent. Simply read the questions and wait for the respondent
to answer. If the person is having problems, the interviewer may encourage
him or her by saying, “Is there anything else?” or “What else did you
do?”
7. When a respondent answers “I don’t know,” misinterprets the meaning of a
question, or contradicts himself/herself, repeat the question or probe by saying,
“Is that all?” or “Could there be other reasons?” or “Could you tell me
some more?”
8. Where it is indicated “Go to question ...”, follow the instruction carefully.
9. Never show that the respondent is wrong when asking questions about his/
her knowledge. Take down answers as given. If the response given indicates
that the person misunderstood, however, repeat the question to clarify.
10. Do not let the respondent go off on a tangent. If he/she talks too long on one
topic, listen for a polite time and then proceed with the interview.
Probing questions. In many interview situations, some respondents tend to give
vague replies such as “okay” or “good,” which can mean different things. When this
happens, try to have the person express himself/herself better by asking why it is
“okay” or “good” and encourage the person to give more specific answers. If a
respondent’s answer belongs to the “other” category of responses in the questionnaire,
ask him/her to specify the response. These follow-up questions are referred to
as probes, which are often used to elicit additional information, expand an idea al-
18 Escalada and Heong

eady expressed by the respondent, or clarify the respondent’s response (Sedlack and
Stanley 1992).
Usually, open-ended probing questions ask for more than a yes or no answer and
provide the respondent the leeway to respond to a question from his/her own perspective.
These generic follow-up questions are suggested to elicit more precise information
(Kidder 1981, Krueger 1988):
“Could you give an example?’
“In what way?”
“What do you mean?”
“Would you explain further?”
“Tell me a little more about it.”
“What do you mean when you said...”
“Tell me how it is so...”
Open-ended survey questions usually provide opportunities for probing, but the
sequence of probe questions to ask depends on the respondent’s initial response (see
the example in the box).
Using probes to get richer information
After establishing that farmers face pest problems and that leaf-feeding insects
are important, the interviewer may want to try to identify the species.
This can be done through a conversation using some of these questions:
• You said previously that ulod 1 are pest problems, can you tell me
what they look like?
• Can you describe the color?
• How big are they?
• Where do they live?
• At what stage of the crop do you see them?
• At what time of day do you see them?
• How many of these insects do you often see on the rice crop?
• What are they doing to the crop?
• Please draw the ulod for me here.
1 Ulod is a term for worm in Cebuano, a local language spoken in most of Visayas and Mindanao
provinces in the Philippines.
Check the responses
1. After an answer is given, check it to make certain it is complete.
2. Review the responses immediately upon completing the interview to check
for any unclear information; if found, clear up the problem with the respondent
before leaving.
3. Correct inconsistent answers by asking the respondent to clarify answers.
Close the interview. After checking your interview schedule for completeness,
briefly thank your respondent for his/her cooperation.
Specific instructions on using the questionnaire. Every questionnaire should have
a set of instructions to guide the interviewer on how certain questions should be asked
Methods for research on farmers’ knowledge, attitudes, and practices in pest management 19

to elicit the needed information. Interviewers should be provided with written instructions
on how to use the questionnaire during their training (see sample below).
Encoding and analyzing survey data. Once the completed questionnaires have
been checked, the data are coded, processed, and analyzed by hand or by using a
statistical package for a computer. Ease of use, power, and cost are some of the important
considerations when choosing computer software for data analysis. Translating
question responses and respondent information into specific categories for analysis is
called encoding the data. The first step of encoding information involves constructing
Sample instructions on using questionnaire
Page
1
Question
no.
Instructions
As you ask the respondent’s name, fill in the name, address,
date of interview, and other information required.
It is important to record this information because we may
want to do a follow-up interview. Recording the
respondent’s name will also facilitate the task of verifying
the interview or the field supervisor in doing spot checks.
1 6
2 7
3 9
3 12-14
Rice yield should be recorded in tons ha -1 . Record the
local unit of measure the respondent gives, but later convert
it to kg and then tons ha -1 .
Read aloud the farm inputs (from 7a to 7g) one at a time
and record the costs incurred for each. For 7g (labor cost
for pesticide application), if the respondent or a family
member provided the labor, ask him/her to estimate how
much he/she would have had to pay a hired laborer to do
the work. If the respondent has difficulty estimating the
cost, assist him/her by asking for the local wage rate for
certain farm operations and multiply this by the number
of days spent performing the task.
Do not read the response categories. Wait for the respondent
to answer. Probe for as many answers as possible
but do not lead or help the respondent with the answers.
After he/she has given an answer, probe by saying, “What
else did you do?”
Most farmers will have applied insecticides several times
during a cropping season. Ask for the times by having the
respondent recall the number of days after planting that
the insecticide was applied. His/her response for the first
application time should correspond with the response in
question 10. Probe for the second time and third time, if
20 Escalada and Heong

a code book, which is a set of rules used to classify observations of variables into
values that are transformed into numbers (see the sample code book below).
The numbers of responses in different categories are tabulated and then statistical
analysis is done, including calculating percentages and averages and doing appropriate
tests of significance. The data analyses are then interpreted and the results organized
into a coherent report that explains the research problem, data collection
methods used, findings, and conclusions. Like other research papers, the survey report
should consist of an abstract or executive summary, introduction, methods, results
and discussion, and conclusions. The abstract. which is a succinct summary of
the work, should describe the survey objectives, methods, summary of the results,
discussion, conclusions, and recommendations.
Sample code book
Q.
no.
Variable
name Column Code Description
Respnum A Enter actual number
1
Variety C 1 IR variety (Hungary, lR29723,
lR13240108, lR50404, IR68, IR64,
IR50401, IR62, IR66, IR65,
lR17433, IR61, IR42, IR60, IR28,
IR250, IR300 595, IR120, lR9729,
lR35546, 9729, 19660, 1352,
2797, 69C, 250, 84-23, 84, 10,
522, 3200, 10-1, 32429)
2 Local variety (F28, F36, G125, G
596, G 02, G 03, G 31851, K, Khac,
Trang Phuoc, Trangnho, NEP)
3 Breeding line from research institutions
(CL, CL6, CL2, CL46, CL950,
CL7, CL5, C28, KSB-54, KSB 55,
DT10, LA-1, LA-70, LM 45, MTL-85,
MTL 59, MTL 54, MTL-58)
2
Sdsource D 1
2
Private seed grower
Ministry of Agriculture
3 Own
4 Neighbor
5 Other (specify)
3
Method E 1
2
Direct seeding
Transplanting
Methods for research on farmers' knowledge, attitudes, and practices in pest management 21

Conclusions
In this chapter, we discussed various methods for determining farmers’ knowledge,
attitudes, and practices in pest management. These methods—rapid rural appraisal,
key informant interview, focus group interview, and formal survey—have been used
by many authors for various purposes to obtain data from the field. The methods are
instruments to gather data and, for the research to provide meaningful descriptions
and analyses, it is important to begin by defining the problem.
At the outset, the research problem has to be defined so that the appropriate
variables can be carefully selected to test the hypotheses; the corresponding instruments
can then be developed to address the problem. Each method has its own strengths
and limitations. Thus, authors often use a combination of methods in their research
program.
Because data from surveys and related methods depend on respondents’ selfreports,
they may contain errors that weaken their reliability and validity. Such errors
may arise from the use of inappropriate variables, biased sampling procedures, wording
of questionnaires, linguistic and cultural nuances, interviewer inadequacies, and
data-encoding mistakes. To minimize such sources of error, special emphasis must be
placed on instrument design and implementation of surveys. It is useful to conduct
several pretests before finalizing the instruments, especially to prevent misunderstandings
caused by linguistic differences. In implementation, it is useful to adopt a quality
assurance procedure to minimize interviewer inadequacies and data encoding.
References
Adhikarya R. 1994. Strategic extension campaign: a participatory-oriented method of agricultural
extension. Rome: Food and Agriculture Organization of the United Nations.
Becker SL. 1970. Rating scales. In: Methods of research in communication. Emmert P. Brooks
WD, editors. Boston: Houghton Mifflin Company. p 213-235.
Bentley J, Andrews K. 1996. Through the roadblocks: IPM and Central American smallholders.
London: International Institute for Environment and Development. Gatekeeper Series No.
56.
Bilston L, Norton G, Frank B. 1997. The impact of participatory problem specification and
planning workshops in pest management. In: Managing change: building knowledge and
skills. Proceedings of the 2nd Australasia Pacific Extension Conference, Albury, New
South Wales.
Bogdan R. Taylor SJ. 1975. Introduction to qualitative research methods. New York: John Wiley
& Sons. p 6-11.
Brader L. 1979. Integrated pest control in the developing world. Ann. Rev. Entomol. 24:225-
254.
Bryman A. 1988. Quantity and quality in social research. London: Unwin Hyman.
Byerlee D, Collinson M, Perrin R, Winkelmann D, Biggs S. 1980. Planning technologies appropriate
to farmers: concepts and procedures. Mexico: Centro Internacional de
Mejoramiento de Maiz y Trigo.
Casley DJ, Lury DA. 1982. Monitoring and evaluation of agriculture and rural development
projects. Baltimore (Md., USA): Johns Hopkins University Press.
22 Escalada and Heong

Chambers R. 1983. Rural development: putting the last first. London: Longman.
Collinson M. 1984. Diagnosing the problems of small farmer needs. In: Davis TJ. editor. Proceedings
of the 4th agricultural sector symposium. New York: The World Bank. p 124-
146.
Conway GR. 1985. Agroecosystem analysis. Agric. Admin. 20:31-55.
Debus M. 1988. A handbook for excellence in focus group research. HEALTHCOM Project
Special Report Series. Washington, D.C.: Porter/Novelli.
Frankfort-Nachmias C, Nachmias D. 1996. Research methods in the social sciences. New York:
St. Martin’s Press, Inc.
Fujisaka S. 1991. A set of farmer-based diagnostic methods for setting post Green Revolution
rice research priorities. Agric. Syst. 36: 191 -206.
Gonzalez AB. 1985. Epilogue. In: Introduction to qualitative research methods. Bautista ML,
Go SP, editors. Manila: De la Salle University Research Center. p 173-184.
Heong KL, Sogawa K. 1994. Management strategies for key insect pests of rice: critical issues.
In: Crop loss assessment in rice. Los Baños (Philippines): International Rice Research
Institute. p 3-14.
Jimenez PR. 1985. Participant observation. In: Introduction to qualitative research methods.
Bautista ML, Go SP, editors. Manila: De la Salle University Research Center. p 144-156.
Kearl B (ed.). 1976. Field data collection in the social sciences: experiences in Africa and the
Middle East. New York: Agricultural Development Council Inc.
Kerlinger FN. 1986. Foundations of behavioral research. New York: CBS College Publishing.
p 377-383.
Kidder LH. 1981. Research methods in social relations. New York: Holt. Rinehart and Winston.
Krueger RA. 1988. Focus groups: a practical guide for applied research. Beverly Hills, CA:
Sage Publications, Inc.
Kumar K. 1987. Conducting group interviews in developing countries. A.I.D. Program Design
and Evaluation Methodology Report No. 8. Washington, DC: U.S. Agency for International
Development.
Merrill-Sands D, Collion MH. 1993. Making the farmer’s voice count: issues and opportunities
for promoting farmer-responsive research. J. FSRE 4(1): 139-161.
Moulton J, Roberts AH. 1993. Adapting the tools to the field: training in the use of focus
groups. In: Seidel RE, editor. Notes from the field: communication for child survival.
Washington, DC: Academy for Educational Development. p 31-37.
Neuman WL. 1997. Social research methods. Boston, MA: Allyn and Bacon.
Norton GA, Heong KL. 1988. An approach to improving pest management: rice in Malaysia.
Crop Protect. 1:84-90.
Norton GA, Mumford JD. 1993. Decision tools for pest management. Wallingford, UK: CAB
International.
Okamura JY. 1985. Ethnographic research methods. In: Bautista ML, Go SP, editors. Introduction
to qualitative research methods. Manila: De la Salle University Research Center. p
157- 172.
Oppenheim AN. 1966. Questionnaire design and attitude measurement. New York: Basic Books.
Inc.
Parel CP, Caldito GC, Ferrer PL, de Guzman GG, Sinaioco CS, Tan RH. 1978. Social survey
research design. Quezon City: Philippine Social Science Council.
Methods for research on farmers’ knowledge, attitudes, and practices in pest management 23

Parel CP, Mijares TA, Orense MM, Samson PQ, Alonzo DC, Concepcion MB, Gutierrez JS,
Tienzo BP. 1966. Introduction to statistical methods (with applications). Manila: Macaraig
Publishing Company, Inc.
Sedlack RG, Stanley J. 1992. Social research: theory and methods, Boston, MA: Allyn & Bacon,
229 p.
Siebert SD. 1973. The integration of fieldwork and survey methods. Am. Sociol. Rev. 78(6): 1335-
1359.
US. Department of Health and Human Services. 1980. Pretesting in health communications:
methods, examples, and resources for improving health messages and materials. Bethesda
(Md., USA): National Cancer Institute.
Yin RK. 1984. Case study research: design and methods. California: Sage Publications.
Notes
Authors’ addresses: M.M. Escalada, Department of Development Communication, Visayas State
College of Agriculture, Baybay, Leyte, Philippines; K.L. Heong, Entomology and Plant
Pathology Division, International Rice Research Institute, Los Baños, Laguna, Philippines.
Acknowledgments: The authors thank the Swiss Agency for Development and Cooperation
(SDC), through the Rice IPM Network based at IRRI, for funding the farmer surveys and
focus group interviews from which we have drawn the content for this chapter, the Food
and Agriculture Organization of the United Nations (FAO) for allowing us to cite some of
its processes and results, and our research collaborators in the national agricultural research
systems in the Philippines, Thailand, and Vietnam.
Citation: Heong KL, Escalada MM, editors. 1997. Pest management of rice farmers in Asia.
Manila (Philippines): International Rice Research Institute.
24 Escalada and Heong

Appendix A
Content Page
Focus Group Interview Report on Integrated Weed Management
in the Muda Irrigation Scheme, Malaysia a
Executive summary
Acknowledgments
Table of contents
Introduction
Objectives of the focus group interview
Methodology
Focus group interview
The respondents
Selecting the respondents
Discussion guide
Introduction
Experiences in rice cultivation
Problems in rice cultivation
Weed problems
Experiences in weed control
Relationship with MADA and other related agencies
Experiences in rice cultivation
Years of involvement in rice cultivation
Changes in rice cultivation
Methods for research on farmers' knowledge, attitudes. and practices in pest management 25

Problems in rice cultivation
High production costs
Irrigation
Marketing and transportation
Pests and diseases
Other crop-related problems
Weed problems in rice cultivation
Extent of weed problem
Types of weeds
Knowledge on weeds
Attitude toward weeds
a Source: Ramli Mohamed and Khor Yoke Lim. 1988. A report of the focus group interview on the strategic
extension campaign on integrated weed management in the Muda irrigation Scheme, Malaysia. Consuitants'
report submitted to the Muda Agricultural Development Authority and FAO.
26 Escalada and Heong

Appendix B
Sample survey questionnaire on knowledge, attitudes, and practices
RICE IPM NETWORK
Respondent
District
Interviewer
Village
Date of Interview
Province
I. BACKGROUND INFORMATION
1. Last cropping season, what rice varieties did you plant?
2. What is your source of rice seeds?
______1) Private seed grower
______2) Ministry of Agriculture
______3) Self
______4) Neighbor
______5) Other (specify)____________________________
3. How is rice grown in your field?
______1) By direct seeding
______2) By transplanting
4. What is your total rice area?________ha
Methods for research on farmers' knowledge, attitudes, and practices in pest management 27

5. What cropping pattern do you follow?
1) Rice-fallow
2) Rice-rice
3) Rice-other crops (specify)
6. What was your rice yield last season? yield per ha
[Note: record actual unit given by farmers and later convert it to kg]
II. PEST MANAGEMENT PRACTICES
7. What pest(s), if any, did you have last cropping season?
1) Armyworm 11) Sheath blight
2) Blast 12) Stem borer
3) Brown planthopper 13) Tungro
4) Caseworm 14) Whitebacked planthopper
5) Gall midge 15) Whorl maggots
6) Green worm 16) Yellowleaf
7) Leaffolder 17) Yellow spot
8) Nematodes
9) Rats
18) Other (specify)
19) None
10) Rice bug
8. What pest, if any, caused the most damage to your rice crop last season?
No. 1
9. What is your second most important pest problem?
No. 2
10. How did you control these pests last season?
10a. Most important pest
1) Apply pesticides (ASK Q. 11)
2) Hand picking
3) Baiting
4) Water management (flooding, draining)
5) Nothing
6) Other (specify)
10b. 2nd most important pest
1) Apply pesticides (ASK Q. 11)
2) Hand picking
3) Baiting
4) Water management (flooding, draining)
5) Nothing
6) Other (specify)
28 Escalada and Heong

11. If you applied pesticides in the last cropping season, how many weeks after planting did
you apply them?
weeks
12. All in all, in the last cropping season, how many times after planting did you apply pesti-
cides?
times
13-16. Please tell me the number of times you applied pesticides at particular
stage(s) of the crop. What chemicals did you apply at these stages? For
which pests? [INTERVIEWER: Make sure that the number of pesticide applications
given in 0. 10 matches the total number of applications reported
for a// stages. You need to probe for pesticides applied at each crop stage
as a farmer could make more than one application at each stage.]
Time of application*
No. of (Q. 14) For which
applications
Pesticides applied pests
(Q. 13)
DAS Crop stage (Q. 15) (Q. 16)
Total
0 Seedbed
1-20 Seedling
21-40 Vegetative
41-60 Panicle initiation
>60 Reproductive
Ripening
* Time of application can be in terms of either days after seeding (DAS) or crop stage.
17. Can you tell me why you sprayed at these times?
Time of application*
DAS Crop stage Reason for pesticide use
0 Seedbed
1-20 Seedling
21-40 Vegetative
41-60 Panicle initiation
>61 Reproductive
Ripening
18. In general, please estimate the percentage of insect pests killed by the insecticides that
you used?
1) 75-100% of insect pests
2) 50% of insect pests
3)

19. How much in total did you spend on chemicals last season?________(local currency)
20. Did you spray the pesticides yourself?
_____1) Yes
_____2) No
20a. IF NO, how much did you pay for labor for spraying?
_____(local currency)
21. Do you have a sprayer?
_____1) Yes
_____2) No
21a. If YES, what kind of sprayer do you have? [INTERVIEWER: Ask the farmer to show
his/her sprayer.]
_____1) Knapsack
_____2) Hand sprayer
_____3) Other (specify)__________________________________
21b. If NO, how do you get hold of a sprayer when needed?
Ill. INFORMATION SOURCES AND KNOWLEDGE ON PEST MANAGEMENT
22. Where did you first hear about the insecticide you were using?
_____1) Neighbor/other farmers
_____2) Extension technician
_____3) Pesticide sales agents
_____4) Mass media
______print
______billboard
______radio
5) Other (please specify)__________________________
23. What is your most important consideration in deciding what insecticide to buy? [INTER-
VIEWER: If respondent gives more than one answer, probe whether those considerations
are most important to him/her before jotting down his/her response.]
_____1) Price
_____2) Effectiveness (killing efficiency)
_____3) Packaging
_____4) Endorsement by sales agent
_____5) Other (specify)
30 Escalada and Heong

24. From where do you get pest control advice?
1) Neighbor
2) Extension technician
3) Relatives
4) Pesticide sales agents
5) Radio
6) TV
7) Other (please specify)
24a. IF MORE THAN ONE RESPONSE IN Q. 24, which of these sources of pest control
advice is most credible to you?
25. Why?
26. Are you aware of any training on pest management conducted in your area?
1) Yes
2) No
26a. If YES, what was the training about?
26b. Who organized the training?
27. Have you attended this training on pest management?
1) Yes
2) No
Knowledge of natural enemies
28. Are there other insects/animals that do not cause damage to your rice crop?
1) Yes (ASK Qs. 28a, 28b, 29)
2) No (GO TO Q. 29)
3) I don't know
28a. If YES, what are they? Name as many as you can.
28b. What do these animals do in your field?
1) Feed on other insects
2) Dwell on leaves or live in the field
3) Don't know
Methods for research on farmers' knowledge, attitudes, and practices in pest management 31

29. What do you think happens to these animals when your rice crop is sprayed with chemicals?
______1) Killed
______2) Disappear
______3) Don't know
______4) Other (specify)
Attitudes toward pesticide use
Please indicate whether you agree, have no opinion, or disagree with the following statements:
30. Applying pesticides to the rice crop will make the yield go up.
______1) Agree
______2) No opinion
______3) Disagree
31. Killing the natural enemies in your rice field by spraying chemicals can cause pest infestation.
______1) Agree
______2) No opinion
______3) Disagree
Knowledge of leaf-feeding insects
32. Which insects that feed on rice leaves do you know?
Please indicate whether you agree, have no opinion, or disagree with the following statements:
33. Leaf-feeding insects (e.g., leaffolder, leafroller) in the early stage cause severe damage
to the rice crop.
______1) Agree
______2) No opinion
______3) Disagree
33a. If YES, please name the leaf-feeding insects that cause severe damage to the rice
crop.
34. Leaf-feeding insects do not cause yield loss.
______1) Agree
______2) No opinion
______3) Disagree
35. Spraying chemicals to control leaf-feeding insects has to be done early.
______1) Agree
______2) No opinion
______3) Disagree
32 Escalada and Heong

IV. SOCIODEMOGRAPHIC PROFILE
36. What is your age? years
37. Please tell me your tenure status:
1) Owner-operator
2) Lessee
3) Tenant
4) Hired laborer
5) Other (specify)
38. What is the highest grade/year in school you have completed?
39. Are you a member of any farmers' organization?
1) Yes
2) No
40. Which farmers' organizations are you a member of?
41. In the last cropping season, did you borrow money for rice production?
1) Yes
2) No
42. From whom did you borrow money?
1) State, how much interest?
2) Private, how much interest?
3) Not applicable
V. COMMUNICATION VARIABLES
43. Do you have a functional radio?
1) Yes
2) No
44. What radio station(s) do you usually tune into?
45. What time do you usually listen to radio?
46. Which types of radio programs do you listen to?
Methods for research on farmers' knowledge, attitudes, and practices in pest management 33

47. Which of the following print materials do you read most?
______1) Newspaper
______2) Comics
______3) Pamphlets
______4) Magazines
______5) Other (specify)______________________________________
48. Do you have a television (TV) set?
______1) Yes
______2) No
49. Which TV station(s) do you watch?___________________________
50. At what time do you usually watch TV?_________________
Thank you.
34 Escalada and Heong

CHAPTER 2
Pest management practices
of lowland rice farmers in Cambodia
G.C. Jahn, P. Sophea, K. Bunnarith, and P. Chanthy
Introduction
We interviewed 1,265 lowland rice farmers in Cambodia on their
pest management practices from June 1995 to April 1996. Fortythree
percent of the farmers were not aware of natural enemies of
rice pests; 59% thought that pesticides increased rice yields. Only
19% believed the insecticide applications could produce pest outbreaks
by killing natural enemies. More farmers applied pesticides
in the dry season than in the wet season, and more men than women
used pesticides. Pesticide users and nonusers did not differ significantly
in average age, education, or farming experience. Nationally,
22% of wet-season farmers and 57% of dry-season farmers used
pesticides. The rice yields of pesticide users and nonusers did not
differ significantly, except in the wet season among farmers who did
not use fertilizers. Insecticides and rodenticides were the most commonly
used pesticides. None of the farmers used fungicides. An
estimated 224,000 liters of pesticides were used annually on rice.
In both the wet and dry seasons, nearly half of those applying pesticides
used a knapsack sprayer. The average reported dry-season
rice yield of 2.5 t ha -1 was significantly higher than the average reported
wet-season yield of 1.4 t ha -1 .
Pest management in Cambodia
Several surveys have been conducted in Cambodia in recent years to address aspects
of determining farmers’ knowledge, attitudes, and practices concerning rice pests.
The first of these, made by the International Rice Research Institute in 1989 (Rapusas
et a 1989), revealed that 28% of the 42 wet-season (WS) rice farmers interviewed in
Takeo Province used insecticides. This survey, however, was conducted in areas where
the government distributed insecticides to farmers, so the sample may have been biased
in favor of insecticide use. Prom (1993) found that 40% of 60 dry-season (DS)
rice farmers surveyed in Kandal Province used insecticides and that 70% believed
insecticides increase yields. Joshi et al ( 1994) reported that of the nearly 100 WS and
35

DS farmers in four provinces interviewed, more than 40% of those in the three provinces
near the Vietnam border used insecticides, but none of those living in the province
not bordering Vietnam used insecticides. Rickman et al (1995) interviewed 197
rice farmers (with no regard to season or ecosystem) in five provinces and found that
10% applied liquid pesticides.
Yech (1994) revealed that methyl parathion was the most popular insecticide
among vegetable and rice farmers in Cambodia. He also found that of the pesticides
being sold in 24 markets in 10 provinces, most belonged to the World Health
Organization’s (WHO) hazard class I 1 . According to Yech (l994), more than 50 pesticides-73%
of which were insecticides—were available in Cambodia representing
30 different active ingredients.
Most pesticides are labeled in Thai or Vietnamese. The only ones with Khmer
labels are the Japanese-produced insecticides fenvalerate, fenitrothion, diazinon, and
cartap (Pisani 1996). Most of the pesticides in Cambodia were found to have far less
active ingredients than indicated on the label (Nesbitt et al 1996).
Integrated pest management (IPM) survey
About 70% of Cambodians are rice farmers (CIAP 1995a). Lowland rice (irrigated
and rainfed) is grown on most of the cultivated riceland in Cambodia; the most important
crop protection issues focus on that crop. Through the Cambodia-IRRI-Australia
Project (CIAP) IPM Program, we conducted a national survey to determine key rice
pests, defined as ones that regularly exceed economic threshold levels (Bottrell l979).
The goal of the study was to document which lowland rice pests farmers consider
important and then focus research on determining the effects of those pests on yields.
No laws or regulations deal with pesticides in Cambodia and anyone can purchase
pesticides—unrestricted—from Thailand and Vietnam (Pisani 1996). Pesticide
use is also unregulated. For example, some merchants spray pesticides on fish sold in
open air markets to repel flies (Chaumeau 1996).
This survey was conducted to determine
• pest management practices of lowland rice farmers.
• attitudes of lowland rice farmers toward pests and pesticides,
• what farmers would like to know about pest management, and
• farmers’ knowledge of the rice ecosystem.
These findings will be used to
• address the pest problems farmers consider as the most pressing,
• evaluate farmer practices scientifically so that recommendations can be made
to farmers and organizations that assist farmers,
• identify dangerous or ineffective pest management practices,
• conduct research to improve on dangerous or ineffective practices,
1 WHO category I pesticides are “extremely and highly hazardous to human health.”
36 Jahn et al

• assist the Cambodian Government in developing training courses for rice farmers,
and
• understand misconceptions.
Survey site
Geographic location
Thailand, Lao PDR, and Vietnam border Cambodia. Four mountain ranges—the
Cardomen and Elephant mountains to the west and southwest, the Dangrek mountains
to the north, and the central highlands of Vietnam to the east—are near Cambodia’s
borders. These mountains surround a vast central plain containing the Tonle Sap lake
and river complex (Nesbitt 1996). The rice-growing areas of Cambodia consist of the
soils developed (from underlying parent material) on the old alluvial or colluvial plains
and on the active floodplains (White et al 1996).
Current state of Cambodian development
Cambodia is one of the poorest and least developed countries in Asia. There is one
telephone for every 1,200 people and more than 9,500 people per doctor. In comparison,
Lao PDR has 300 people per telephone and 4,400 people per doctor. Only 38%
of Cambodia’s population is literate, whereas literacy rates exceed 80% in Lao PDR,
Thailand, and Vietnam. Of the four countries, Cambodians consume the least calories
per day. Cambodia has the highest infant mortality rate in Southeast Asia, with 11%
of Khmer babies dying before turning one year old. Nevertheless, Cambodia’s population
is growing by 2.5% per year (Asiaweek 1996).
Rice production in Cambodia
From 1960 to 1969, the area planted to rice in Cambodia averaged 2.4 million ha per
year (Khush et al 1986). Cambodian rice production peaked in 1969-70 at 4 million t
(Helmers 1996, MAFF 1996). From 1970 to 1975, rice production declined drastically
as the country plunged into civil war. The area cultivated to rice decreased by
77% between 1970 and 1974, and production dropped 84% (Hildebrand and Porter
1976). In 1975, the Khmer Rouge took over Cambodia. For the next four years, its
members attempted to create a classless, agrarian, pure Khmer society. To boost rice
production, the Khmer Rouge forced almost the entire population to grow crops or
build irrigation canals. But the Khmer Rouge’s efforts to increase rice production
were generally unsuccessful (Watts et al 1989). Vietnam succeeded in removing the
Khmer Rouge from power in 1979 and pushed its forces to the Thai border. The
regime continues to control territory in western Cambodia. In 1979, only 770,000 ha
of rice were sown and a mere 588,000 ha were harvested, producing about 565,000 t
of grain (FA0 1980, MAFF 1996). As a result, 35% of Cambodia’s 1980 food needs
were supplied from international sources (Mason and Brown 1983).
The area under rice has grown steadily since 1979: 1.4 million ha in 1987, 1.7
million ha in 1992, and 2 million ha in 1995 (CIAP 199 1, 1995b, MAFF 1996). About
1.6% of the potential lowland rice area cannot be used because of land mines (Nesbitt,
Pest management practlces of lowland rice farmers In Cambodla 37

unpublished data). Cambodian rice production reached 3.3 million t in 1995—about
the same production level as achieved in 1967 on 2.5 million ha (CIAP 1991, 1995b,
MAFF 1996). The major difference is that in 1967 the population was 6.6 million,
whereas in 1995 it was 10 million (Asiaweek 1996, Tichit 1981).
Rice ecosystems in Cambodia
Wet-season rice. Most (93%) Cambodian riceland is classified as lowland, and the
rest (7%) as upland and deepwater. Most of Cambodia's lowland rice is grown during
the wet season. Rainfed lowland rice (RLR) accounts for 88% of Cambodia's cultivated
riceland, while the remaining 5% is irrigated. Most of this land is planted to
late-maturing varieties, some to medium-maturing varieties, and limited amounts to
early maturing varieties (Lando and Mak 1994a). Late-maturing varieties are usually
planted in low fields where the water levels frequently exceed 50 cm. Medium-maturing
varieties are usually planted in middle fields, where the highest standing water
levels seldom exceed 40 cm, and in high fields, where the water levels seldom exceed
20 cm. Early maturing varieties are commonly planted in high fields (Lando and Mak
1994b). Roughly 90% of WS rice seed is sown in May through July, and 60% is
harvested in December (FAO 1996).
Dry-season rice. Since 1982, DS rice has accounted for no more than 5-8% of
the total rice area cultivated nationally (Lando and Mak 1991). It is not known what
portion of DS land is used for two or more crops per season. The ability to grow DS
rice depends on the presence of water, generally from wetlands and rivers. More than
30% of Cambodia is considered wetland (Selvanathan 1993), suggesting considerable
potential to expand DS rice cultivation. DS rice seed is sown from late October to
mid-November and harvested in March. Nearly all DS land is planted to early maturing
varieties (Javier 1996). The modern, photoperiod-insensitive varieties can be grown
virtually any time of the year, as long as the drainage and irrigation system is adequate.
For example, in parts of Suay Rieng Province near Vietnam, farmers plant an
early maturing irrigated crop from May to July.
At least 15 orders of arthropods inhabit Cambodian lowland rice. Most of these
are not herbivores or plant feeders (Arida and Banion 1995, Jahn et al 1996).
Rice research in Cambodia
Agricultural research ceased in 1970 because of the government's preoccupation with
the civil war. During the Khmer Rouge regime, nearly all scientists were killed or fled
the country, educational institutions were closed, and research libraries and government
publications were burned. With the exception of some papers published in French
journals before 1953, nearly all Cambodian agricultural research documentation was
destroyed (CIAP Khmer staff members, pers. comm.).
Rice research began again in 1986 when IRRI and the Cambodian Ministry of
Agriculture, Forestry, and Fisheries (MAFF) signed a Memorandum of Understanding.
Since 1987, the Australian Agency for International Development (AusAID) has
financed this collaborative rice research effort, known as CIAP (CIAP 1994).
38 Jahn et al

Methods
We collected data on farmers’ pest management knowledge, attitudes, and practices
through structured personal interviews using a questionnaire. The survey instrument
was based on one used in Vietnam to test the effects of a no-early-spray campaign
(Heong et al 1994, Rapusas et al 1994). We did five pretests using 25 different farmers
each time.
Pretesting revealed one serious obstacle. Farmers were originally asked to name
the most common pests, but little agreement existed on local names of arthropodseven
within the same village. To overcome this problem, we created a poster of randomly
mixed photographs and drawings of common rice insect pests and natural enemies,
weeds, and diseases. Instead of providing names, farmers were asked to point
to the pests that caused problems in their fields. They were then asked if there were
other pests not shown on the poster.
The questionnaire was originally written in English and then translated into Khmer.
Fourteen Cambodians, trained in survey techniques and in using the questionnaire,
served as the interviewers.
From June 1995 to April 1996, we interviewed 1,265 farmers from 154 villages
in 10 provinces—Battambang, Kampong Cham, Kampong Chhnang, Kampong Speu,
Kandal, Prey Veng, Pursat, Siem Reap, Svay Rieng, and Takeo (Fig. 1). These provinces,
out of the country’s 21, represent 76% of the total lowland rice area. Survey
sites were chosen on the basis of accessibility, security, and lowland rice production.
In each village, a group of two to five interviewers conducted the survey. No lists of
the residents of the villages existed, so the interviewers simply chose farmers for the
survey. They were, however, instructed to include both men and women of different
economic status and of a variety of ages. A given farm was only represented once in
the survey.
While not truly a random sample, the survey results do not appear to represent
any bias in farmer selection on the part of the interviewers. Given that parametric
tests are robust, we assumed a normal distribution and a random sample for the purpose
of data analysis.
The survey data were coded and entered into the Microsoft Excel ® spreadsheet
(Version 5.0) on a computer. Data were analyzed with the Microsoft Excel ® and
Statistica ® software programs (Microsoft Corporation 1993). Averages were compared
by t-tests, and frequency associations were measured by chi-square tests.
Pest management practices of lowland rice farmers in Cambodia 39

Fig. 1. Map of Cambodia. Each dot represents one of the 168 villages included in the IPM survey.

Survey results and discussion
Farmer profile
Fifty-seven percent of the respondents were men and 43% women. These percentages
should not be taken to indicate the sex ratio of rice farmers because we made no
attempt to record the number of men and women in each household. Their ages ranged
from 17 to 76 years and averaged 43 years. Respondents had 0- 13 years of education,
with the average being 3.7 years. Most had spent the greater part of their lives as rice
farmers, with their farming experience ranging from 1 to 60 years and averaging 23
years. Since the 1993 elections, land tenure remains an unresolved issue in Cambodia;
nonetheless, 99% of WS farmers and 100% of DS farmers claimed that they
owned their largest fields.
Wet season. Families growing WS rice had 1-14 fields; the average was three.
Field size ranged from 0.05 to 7.0 ha, with an average of 0.6 ha. About 77% of WS
fields are never flooded by river water, 22% are flooded once a year, and 1 % twice a
year. Most farmers said they grow traditional varieties. Only 1% of WS farmers reported
growing IRRI-derived varieties, representing an estimated 0.9% of the national
area planted to rice.
Dry season. Families planting DS rice had one to seven fields, with an average of
two. The DS fields ranged from 0.05 to 3.5 ha, with an average of 0.5 ha. Ninety-two
percent of these DS fields are flooded with river water once a year; the rest are not.
Most DS farmers (82%) grow IRRI-derived varieties, representing 92% of the national
DS rice area. This high percentage of IRRI-derived varieties used in the Cambodian
DS is consistent with the findings of Joshi et al (1994).
Farmer knowledge of pests and natural enemies
Nationwide, farmers named more than 50 different pests they consider important in
rice. Except for umbrella sedge ( Cyperus difformis ) in the DS, no particular pest was
cited by more than half the respondents. The ones most commonly reported were
umbrella sedge, rats, yellow leaf, stem borers, and jungle rice ( Echinochloa colona ).
Yellow leaves can be caused by any number of diseases or nutrient deficiencies. Tungro
disease can cause yellow leaves, although its vector, the green leafhopper, was rarely
reported as a pest. More than 20% of WS farmers considered crabs to be a major pest,
but less than 10% of DS farmers did. Joshi et al (1994) also found farmers to frequently
mention stem borers, rats, and crabs as major pests.
More than 30% of DS growers considered caseworms to be a serious pest, whereas
only 17% of WS growers did (Table 1). Of the farmers reporting caseworm problems,
one-third of WS farmers and more than 70% of DS farmers attempt to manage them.
Ironically, caseworms should be easier to manage in the dry season when farmers
have more control over the water in their fields. Draining ricefields provides good
control of caseworms in Cambodia, with damage occuring only in those parts of the
field with stagnant water (Kun 1994). DS farmers, however, mainly use pesticides on
caseworms. Only one DS farmer reported controlling them by draining water from
the field.
Pest management practices of lowland rice farmers in Cambodia 41

Table 1. Pests most commonly reported by Cambodian lowland rice farmers.
Wet season
Pest
Type of
pest
Farmers Of those Most common
reporting reporting, % control
(%) controlling method
the pest
Umbrella sedge
Rats
Whitehead
(stem borer damage)
Stem borer
Crabs
Jungle rice
Caseworm
Weed 39.3
Mammal 27.2
Insect 23.9
Insect 19.2
Crustacean 20.8
Weed 20.4
Insect 17.4
94.7 Removal by hand
33.5 Rodenticide
11.7 Removal by hand
10.9 Removal by hand
56.8 Removal by hand
97.5 Removal by hand
26.0 Ashes
Dry season
Pest Type of Farmers Of those Most common
pest reporting reporting, % control
(%) controlling method
the pest
Umbrella sedge Weed 66.2 85.5
Rats Mammal 45.9 59.7
Whitehead
(stem borer damage) Insect 31.3 30.7
Stem borer Insect 18.5 65.4
Crabs Crustacean 9.6 48.0
Jungle rice Weed 31.3 93.3
Caseworm Insect 30.2 67.6
Removal by hand
Rodenticide
Insecticide
Insecticide
Insecticide
Removal by hand
Insecticide
Farmers are generally aware of pest damage to their rice but do not necessarily
understand the cause. For example, more farmers reported stem borer damage than
reported stem borers (Table 1).
About 43% of farmers were unaware that natural enemies exist, and some reported
natural enemies as their major pests. The most extreme case of this was in
Kandal, where 43% of DS farmers reported ladybird beetles as pests. Besides eating
insects, ladybird beetles do in fact eat rice pollen, but there is no evidence that this
affects yield. Except for ladybird beetles, less than 8% of farmers reported any given
species of natural enemy as a major pest in any province in any season (Table 2).
Only 1 % of farmers mistakenly thought that some pests, such as crabs and mole
crickets, do not damage rice. On the other hand, 87% of farmers knew that some
animals in the field do not damage rice; frogs were cited the most often. Of these
farmers, 66% knew that frogs and other predators eat insects, and more than half of
the 66% thought that pesticides could kill natural enemies.
Attitudes toward pesticide use
Farmers were asked to express their agreement or disagreement with several attitude
statements. The majority (59%) agreed with the statement "Applying pesticides to
42 Jahn et al

Table 2. Natural enemies reported as major pests by Cambodian lowland rice
farmers.
% reporting as % reporting as
Natural enemy Type a wet-season a dry-season
pest
pest
Carabid beetles Predators of 5.9 0.0
insect larvae
Spiders Predators of 5.7 1.4
most insects
Ladybird beetles
Predators of
Homopterans 5.3 2.8
Long-horned grasshopper Predators of 5.2 6.7
( Conocephalus stem borer
longipennis )
eggs
rice increases yields,” whereas 36% disagreed, 4% said they did not know, and the
rest said it depends on the situation. This is consistent with Prom (1993), who also
found that most Cambodian rice farmers thought that pesticides increase rice yields.
Among the farmers aware of natural enemies, 33% agreed that killing natural
enemies with pesticides can cause pest outbreaks, 45% disagreed, and 22% said they
did not know. Overall, only 19% of the farmers interviewed thought that pesticide
applications could produce pest outbreaks by killing natural enemies.
Pest management practices
Pesticide use. The percentage of farmers using pesticides varied greatly from one
province to another: 8-50% of WS farmers, and 40-100% of DS farmers. Nationally,
27% of farmers used pesticides in the WS, and 59% in the DS. Among those applying
pesticides in the WS, 20% used insecticides, 8% used rodenticides, and 1% used
herbicides. Among those applying pesticides in the dry season, 41% used insecticides,
25% used rodenticides, and 1% used herbicides (Table 3).
Like Yech (1994), we found that methyl parathion is the most commonly used
insecticide. Interestingly, methyl parathion is one of the few pesticides sold in Cambodian
markets that contains as much active ingredient as claimed on the label (Nesbitt
et a1 1996). Our findings agreed with those of Yech (1994) that zinc phosphide is the
only rodenticide used in Cambodia. Herbicide users did not know what kind of herbicide
they used, although 2,4-D, paraquat, and alachlor are the ones available in Cambodian
markets.
Herbicide use was highest in Battambang, which is consistent with findings of
Rickman et a1 (1995). Rapusas et a1 (1989) found no herbicide use in Takeo (Table 3).
None of the farmers interviewed used fungicides, though some erroneously applied
insecticides for fungal diseases such as brown spot.
Pesticides were the most common method of controlling major DS pests except
for weeds. Among the farmers using insecticides, more than 93% looked for insect
damage to decide when to apply insecticides. The other farmers sprayed on a sched-
Pest management practices of lowland rice farmers in Cambodia 43

Table 3. Pesticide use in Cambodian lowland rice by season.
Pesticide Pesticide Pesticide
Farmers Using users users users
Province interviewed pesticide applying applying applying
(no.) (%) insecticide rodenticide herbicide
(%) (%) (%)
Wet season
Battambang
Kampong Cham
Kampong Chhnang
Kampong Speu
Kandal
Prey Veng
Pursat
Siem Reap
Svay Rieng
Takeo
Dry season
Kampong Cham
Kampong Chhnang
Kampong Speu
Kandal
Prey Veng
Siem Reap
Svay Rieng
Takeo
119
100
125
104
108
109
115
105
111
107
8
31
42
23
41
71
13
52
37
16
8
11
20
12
14
33
50
13
88
52
40
65
71
49
100
54
59
69
70
82
36
100
69
6
100
50
100
44
94
73
100
3
100
82
34
44
30
27
82
0
25
100
0
64
86
56
18
53
14
100
0
43
27
13
20
0
5
0
6
0
0
0
29
0
0
0
0
0
0
0
ule or simply copied their neighbors’ spraying schedules. Of those who looked for
pests or damage, 73% applied insecticides whenever they saw pest damage, and 26%
sprayed when they observed a certain number of insects (regardless of type). The
remaining 1% of farmers sprayed any insect, only sprayed certain kinds of insects, or
sprayed when pests exceeded predators.
The 10 provinces in this survey have 1.3 million ha of WS lowland rice and
159,000 ha of DS lowland rice (Javier 1996). Assuming that 20% of WS farmers and
41% of DS farmers used insecticides and that the same percentage of rice received
insecticides, then about 260,000 ha of WS rice and 65,000 ha of DS rice received
insecticide applications. Farmers were not usually certain how many times they applied
insecticides in a season, but knew at which stages of the rice crop they applied
them. By counting the number of stages in which they applied pesticides, we were
able to determine the minimum number of insecticide applications in a season for
each farmer. In the DS, farmers applied an average of 0.7 liters ha -1 of insecticide an
average of 2.7 times. We therefore estimated that DS farmers (in these 10 provinces)
applied at least 123,000 liters of insecticide. WS farmers applied an average of 0.5
liters ha -1 of insecticide at least 1.3 times, for a total of 169,000 liters. Thus, we estimate
that at least 292,000 liters of insecticide were applied annually to the lowland
rice in the 10 provinces surveyed.
It appears that knapsack sprayers are becoming more popular in Cambodia. In
1989, not a single WS lowland rice farmer of the 42 interviewed in Takeo used a
44 Jahn et al

knapsack sprayer (Rapusas et al 1989). In contrast, we found that 3% of the WS
farmers in the province (107 respondents) and 19% of the DS lowland farmers (52
respondents) used knapsack sprayers. Nationally, 10% of the 1,103 lowland WS farmers
interviewed and 27% of the 281 lowland DS farmers used knapsack sprayers. This
means that 48% of the WS and DS lowland farmers applying pesticides used knapsack
sprayers. Rickman et al (1995) found that 2% of 197 farmers across ecosystems
used knapsack sprayers.
In this survey, 3% of pesticide users reported pouring insecticides into bowls and
flicking the chemicals over their fields with brushes or leaves. About 9% used a homemade
plunger-type sprayer that squirts large volumes of insecticide over a few spots
in the field. These plunger-sprayers consist of one metal or bamboo tube inside of
another. They are usually used for applying insecticides to seedbeds. The rest of the
farmers mix pesticides with bait and place it in the field for rat control. Only 6% of
DS farmers and 1% of WS farmers who use pesticides reported that they regularly
mix different pesticides together. A few farmers mixed pesticides in the same container
they used to fetch drinking water or to feed domestic animals.
Cultural and physical control. Farmers reported using mainly hand weeding to
control weeds. Diseased or pest-damaged rice plants are commonly pulled from fields.
According to Rickman et al (1995), farmers spend an average of 15.6 d ha -1 hand
pulling weeds from fields each season.
Botanical pest control. Several farmers mentioned that they chop up a cactus-like
plant and place it in the water to kill or repel crabs. A few farmers reported that when
they see yellow rice plants in the field, they stick Chromolaena odorata branches into
the ground upright to make a fence around the yellow rice, which then turns green
within a week. C. adorata, called kanthraing khait in Khmer, is usually not cultivated.
Joshi et a1 (1994) reported that some farmers chop up the leaves of C. orlorata,
eucalyptus, or papaya and broadcast them in the ricefield to drive away crabs or insects.
Eucalyptus oils are known to repel ants (Jahn 1991). but the insect repellent
properties of papaya and C. odorata are unknown. Surprisingly, no one has reported
Cambodian rice farmers using the neem tree (Azadirachta indica) for pest control,
although some farmers use neem leaves to help preserve stored grain. The neem tree
is common in the country and neem parts are used in the traditional Khmer diet and
medicine. Neem leaves and extracts are well documented to have adverse effects on
many insects (Lim and Bottrell 1994).
Unusual pest management practices. Several farmers reported that they treat pest
problems by applying handfuls of fertilizer, ashes, or salt to the damaged area. Many
use smoke to drive away rice bugs and other insects. Some farmers unwind old videotape
and encircle their fields with it to keep the birds away. As the tape flutters in the
wind and reflects sunlight, it scares the birds. Some farmers mix fish oil with crushed
rice bugs to attract other rice bugs to traps (Rapusas et al 1989).
information transfer
Many farmers (60%) reported receiving their pest management advice from their neighbors,
whereas only 22% obtain advice from extension workers (Fig. 2). One farmer
Pest management practices of lowland rice farmers in Cambodia 45

Fig. 2. Sources of pest control advice for lowland rice farmers in Cambodia.
Some farmers receive advice from more than one source; therefore,
the sum of percentages exceeds 100%. N = neighbors, Ra = radio,
E = extension, TV = television, P = pesticide seller, Re = relatives.
said he did not seek advice—instead, he performs his own experiments to see what
works best.
Farmers were asked what subjects they would like to know more about. About
35% had questions about pests, 30% wanted to know about fertilizers, 20% had no
questions, 18% had questions about pesticides, and the remainder were interested in
farm management. When asked how they would like to get this information, 45% said
in school or a training program, 23% reported radio, 23% mentioned pamphlets or
magazines, 7% indicated TV, and 1 % specified extension. Some respondents mentioned
more than one preferred means of receiving information. About 6% expressed
no interest in receiving additional information. The relatively strong desire to learn in
school or attend a training program suggests that such programs could have a significant
impact in Cambodia.
Relationships and patterns in pest management
Yield and pest control. Farmers were asked how much rice they harvested from their
largest field in the previous season. National average DS yields of 2.5 t ha -1 were
significantly higher than average WS yields of 1.3 t ha -1 (P

Table 4. Lowland yield differences in relation to pest control and fertilizer use a .
Dry season
Wet season
n Mean P SD n Mean P SD
yield
yield
(t ha -1 ) (t ha -1 )
Pest control 270 2.4

Table 7. Lowland yield differences in relation to pesticide among pesticide users and nonusers.
Dry season
Wet season
Mean
Mean
n yield P SD n yield P SD
(t ha -1 ) (t ha -1 )
Pest control 181 2.5 0.43 1.49 351 1.1

I
Implications for research
Rice pest constraints. Farmers named more than 50 pests that cause problems in their
ricefields. With the exception of umbrella sedge, no single insect pest or disease was
cited as a problem by more than half the farmers in either season. To determine which
pests named by farmers actually affect yields negatively, we are conducting a rice
pest constraint study. We are collecting and identifying pests from five DS and 15 WS
ricefields four times per season for four years. By relating the information on pest
abundance in each field to the yields in each field, we will detect which pests suppress
yields. Tungro-like symptoms were reported in both the WS and DS, which led to a
search for tungro in Cambodia. IRRI virologist Ossmat Azzam, through immunological
tests, has positively determined the presence of tungro virus, although at low levels.
Effects of pest management practices on yields. We will conduct the same analysis
on the rice pest constraint yield data that we conducted on the survey yield data to
see whether the relationship between fertilizers, pesticides, and yield is the same. In
addition, we will conduct replicated field trials comparing the yields of untreated
plots, plots treated with pesticides by farmers following their usual practices, and
plots treated with pesticides based on the economic thresholds described by Reissig
et a1 (1985).
Botanical control. More than 20% of WS farmers consider crabs a pest problem.
Some farmers reported that placing pieces of a cactus in the water prevents crab damage.
We are conducting experiments to see whether this technique has merit.
References
Arida GS, Barrion AT. 1995. Cambodia trip: accomplishment report. Cambodia-IRRI-Australia
Project. 22 p.
Asiaweek. 1996. Vital signs. May 3, 1996:60.
Bottrell D. 1979. Integrated pest management. Washington (D.C., USA): Council on Environmental
Quality. 120 p.
Chaumeau C. 1996. “Good” insects seen as alternative to chemicals. Phnom Penh Post, May
17-30, 1996:16.
(CIAP) Cambodia-IRRI-Australia Project. 1991, Rice production in Cambodia. Phnom Penh
(Cambodia): CIAP. 45 p.
(CIAP) Cambodia-IRRI-Australia Project. 1994. Facts about cooperation: Cambodia and IRRI.
Phnom Penh (Cambodia): CIAP. 14 p.
(CIAP) Cambodia-IRRI-Australia Project. 1995a. The Cambodia-IRRI-Australia Project:
programmes and achievements. Phnom Penh (Cambodia): CIAP. 18 p.
(CIAP) Cambodia-IRRI-Australia Project. 1995b. Annual report for 1994. Phnom Penh (Cambodia):
CIAP. 251 p.
(FAO) Food and Agriculture Organization. 1980. Report of the FAO Food Assessment Mission,
Office for Special Relief Operations, Kampuchea. November 1980. Rome (Italy):
FAO.
(FAO) Food and Agriculture Organization. 1996. 1995 rice crop pilot survey technical report.
Phnom Penh (Cambodia): Ministry of Agriculture, Forestry, and Fisheries; Department of
Planning and Statistics. 31 p.
Pest management practices of lowland rice farmers in Cambodia 49

Helmers K. 1996. Rice in the Cambodian economy: past and present. In: Nesbitt HJ, editor.
Rice production in Cambodia. Phnom Penh (Cambodia): University Press. p 1-25.
Heong KL, Escalada MM, Mai V. 1994. An analysis of insecticide use in rice: case studies in
the Philippines and Vietnam. Int. J. Pest Manage. 40(2): 173-178.
Hildebrand G, Porter G. 1976. Cambodia: starvation and revolution. New York (New York):
Monthly Review Press.
Jahn GC. 1991. Ant repellent activity of eucalyptus extracts in choice tests. Insect. Acaric.
Tests 16:293.
Jahn GC, Bunnarith K, Sophea P, Chanty P. 1996. Pest management in rice. In: Nesbitt HJ,
editor. Rice production in Cambodia. Phnom Penh (Cambodia): University Press. p 134-
147.
Javier EL. 1996. Rice ecosystems and varieties. In: Nesbitt HJ. editor. Rice production in Cambodia.
Phnom Penh (Cambodia): University Press. p 72-1 33.
Joshi RC, Guino RA, Elliot PC, Preap V, Khiev B. 1994. Informal surveys on farmers’ knowledge,
attitudes, and practices on crop and pest management in Cambodia. Cambodia-
IRRI-Australia Project.
Khush GS, Puckridge DW, Denning GL. 1986. Trip report to the People’s Republic of
Kampuchea, January 23-30, 1986. Manila (Philippines): IRRI. 38 p.
Kun L. 1994. Measurement of yield losses due to insects and other pests at Day Eth station.
B.S. thesis. The Royal University of Agriculture, Phnom Penh. 68 p. (In Khmer.)
Lando RP, Mak S. 1991. Srey Ampal: a baseline survey of dry season rice in Cambodia. CIAP
Baseline Survey Report No. 4. 100 p.
Lando RP, Mak S. 1994a. Rainfed lowland rice in Cambodia: a baseline survey. IRRI Research
Paper Series No. 152.20 p.
Lando RP, Mak S. 1994b. Cambodian farmers decision making in the choice of traditional rainfed
lowland rice varieties. IRRI Research Paper Series No. 154. 17 p.
Lim GS, Bottrell DG. 1994. Neem pesticides in rice: potential and limitations. Los Baños
(Philippines): IRRI. 69 p.
(MAFF) Ministry of Agriculture, Forestry, and Fisheries, Kingdom of Cambodia. 1996. Food
security in Cambodia: a country position paper. World Food Summit, Regional Conference
in Western Samoa, May 1996. Phnom Penh (Cambodia): MAFF. 17 p.
Mason L, Brown R. 1983. Rice, rivalry, and politics: managing Cambodian relief. Notre Dame
(Indiana): University of Notre Dame Press. 218 p.
Microsoft Corporation 1993. User’s Guide: Microsoft Excel Version 5.0. Cambridge (Mass.,
USA): Microsoft Corporation. 784 p.
Nesbitt HJ. 1996. Topography, climate and rice production. In: Nesbitt HJ, editor. Rice production
in Cambodia. Phnom Penh (Cambodia): University Press. p 26-35.
Nesbitt HJ, Hickey A, Phirun I. 1996. Adulterated pesticides in Cambodia. Int. Rice Res. Notes
21(1):51.
Pisani E. 1996. Pesticides poisoning Cambodia’s earth. Asia Times, March 18, 1996.
Prom T. 1993. The practices of farmers for preventing and controlling insects. B.S. thesis. The
Royal University of Agriculture, Phnom Penh. 30 p. (In Khmer.)
Rapusas HR, Bandong JB, Heong KL. 1989. Trip report to Cambodia: October 10-November
28, 1989. Manila (Philippines): IRRI. 64 p.
Rapusas HR, Dedolph C, Escalada MM, Heong KL. 1994. Workshop report: message design
for a campaign to encourage farmers’ participation with stopping early insecticide spraying
in South Vietnam. Los Baños (Philippines): International Rice Research Institute.
146 p.
50 Jahn et al

Reissig WH, Heinrichs EA, Litsinger JA, Moody K, Fiedler L, Mew TW, Barrion AT. 1985.
Illustrated guide to integrated pest management in rice in tropical Asia. Manila (Philippines):
IRRI. 41 1 p.
Rickman JF, Meas P, Om S, Khiev CT, Nhem S. 1995. Farm mechanization and crop production
in Cambodia. CIAP Baseline Survey Report No. 5. 20 p.
Selvanathan S. 1993. Wetland conservation and fisheries management in Cambodia. In: Environment
and integrated pest management seminar, January 11-14, 1993, Chamcar Daung
Agricultural Institute. Phnom Penh (Cambodia): Phnom Penh Post Publishing, Ltd. p 2-3.
Tichit L. 1981. Lagriculture au Cambodge. Phnom Penh (Cambodia): Lagence de Cooperation
Culturelle et Technique. 424 p.
Watts K, Draper C, Elder D, Harrison J, Higaki Y, Salle J. 1989. Report of the Kampuchea
needs assessment study. Phnom Penh (Cambodia): United Nations Development
Programme. 277 p.
White PF, Oberthur, Sovuthy P. 1996. Soils and rice. In: Nesbitt HJ, editor. Rice production in
Cambodia. Phnom Penh (Cambodia): University Press. p 36-57.
Yech P. 1994. A study on pesticides in Cambodia and use habits of Cambodian farmers. B.S.
thesis. The Royal University of Agriculture, Phnom Penh, Cambodia. 32 p. (In Khmer.)
Notes
Authors’ addresses: G.C. Jahn, International Rice Research Institute, Cambodia-IRRI-Australia
Project, P.O. Box 1, Phnom Penh, Cambodia; P. Sophea, K. Bunnarith, and P. Chanthy,
Department of Agronomy, Ministry of Agriculture, Forestry, and Fisheries (MAFF), 14
Monireth Street, Phnom Penh, Cambodia.
Ackowledgments: We thank the Ministry of Agriculture, Forestry, and Fisheries, and the provincial
agricultural offices for helping to coordinate visits to remote villages, particularly
in areas where there were security risks. Ms. Chhem Chantha, Mr. Hor Sophal, Mr. Khiev
Chan Theavy, Mr. Khuon Samoeun, Mr. Men Chhon, Mr. Nhem Sokha, Mr. San Vuthy,
Ms. Seung Keo Viseth, Mr. Som Sarun, Mr. Soun Kimsan, and Ms. Suy Sakunthea of the
Department of Agronomy and the provincial agricultural offices helped conduct interviews
and we are grateful for this assistance. We also thank the staff of PRASAC 3 in Svay
Rieng for conducting interviews.
We appreciate the helpful editorial comments of Dr. K.L. Heong, Ms. Carolyn Dedolph,
and Dr. M.M. Escalada. Special thanks go to Dr. Harry Nesbitt for access to unpublished
data on land mines; and to Mr. Joseph Rickman, Mr. Kent Helmers, Dr. Peter White, Dr.
Edwin Javier, and Dr. G.S. Sidhu for useful suggestions on the survey format and data
analysis. This study was a collaborative effort among the Cambodian Department of
Agronomy, IRRI, and the Australian Agency for International Development (AusAID);
the Government of Australia, through AusAID, funded the research.
Citation: Heong KL, Escalada MM, editors. 1997. Pest management of rice farmers in Asia.
Manila (Philippines): International Rice Research Institute.
Pest management practices of lowland rice farmers in Cambodia 51

CHAPTER 3
Pest management practices
of rice farmers in Hunan, China
Li Shao-Shi, Guo Yu-Jie, Hu Guc-Wen, and Liang Di-Yun
A survey of rice farmers’ knowledge, attitudes, and practices of integrated
pest management was conducted in Wangcheng County, Hunan
Province, China, in 1991. Two hundred sixty-six farmers were interviewed.
Most (93%) of the farmers believed they could not control
insect pests without using pesticides, which they applied 9.9 times
per year on average. Seven percent of them believed beneficial insects
could keep pest populations from developing, but only 8% knew
about parasitoids. Many (62%) obtained pest management information
from the local technology extension station. But when actually
making a pest control decision, about half relied on their own experience
or that of a neighbor’s. Two-thirds of the farmers had not
attended any pest management training.
Introduction
To improve pest management in rice in China, emphasis has been placed on developing
new control techniques and pest forecasting methods (Du 1991). Although many
advances have been made through these efforts, large gaps still exist between research
findings and farmers’ practices. Studies done by the Institute of Agricultural
Economics, Chinese Academy of Agricultural Sciences, revealed that farmers can use
only about 30% of the research findings in agriculture because of constraints in the
interactions among research, extension, and farmers (Zhu 1992). To rectify this situation,
a program for enhancing agricultural development has been launched to make
the results of scientific research—including those of integrated pest management (IPM)
studies-available to farm households. An important first step in implementing IPM
is to understand what farmers think, perceive, and practice.
Researchers have carried out different types of surveys in China (Xu 1989). It
was not until this survey, however, that farmers’ decision making and their knowledge,
attitudes, and practices in rice pest management in Hunan Province were studied,
such as in the surveys conducted by Litsinger et al (1980), Heong (1984), and
Escalada and Heong (1993). Information gathered through these surveys can be use-
53

Fig. 1. Location of survey sites in Wangcheng County, Hunan Province.
ful for identifying gaps between what is known in research and farmer’s practices so
that appropriate interventions can be developed to improve farmers’ pest management
decision making.
Survey site
Hunan Province, which is south of the middle part of the Yangtze River, is one of the
largest provinces in China, where farmers grow a double crop of hybrid rice. Of the
province’s 2.6 million ha of riceland, 21% is planted to hybrid varieties in the first
cropping season and 83% in the second cropping season. Yields average 5.8 t ha -1
season -1 (Hunan Yearbook 199 1).
Wangcheng County, which is near the provincial capital of Changsha (Fig. l),
was chosen as the survey site because it reflects the general conditions in Hunan
Province, including having an efficient irrigation system.
The county's population in 1990 was 757,400, with 90% working as farmers.
The mean annual temperature is 16.4 °C and the mean annual precipitation is 1,500
mm.
The county’s total rice area is 37,700 ha, accounting for 91% of the cultivated
land. Most of the varieties planted are bred in Hunan Province and are high-yielding.
Generally, indica Xianzhao series and the hybrid Weiyou are used for the first crop,
whereas indica varieties Yuchi 231-8, Xiangwan 1. and hybrids Weiyou 64, Weiyou
35, and Xianyou 64 are planted for the second crop.
54 Shao-Shi et al

Methods
We interviewed individuals for the survey. The questionnaire contained 42 questions,
and was pretested on 42 farmers and revised five times by subject-matter specialists
and extension workers in the Plant Protection Center of Wangcheng County.
Agrotechnicians, who were trained for one day on the survey, carried out the
interviews in April-May 1991. They randomly selected 266 farm households in 11
villages distributed in nine townships for the survey. Data were analyzed at the Crop
Pest Forecasting Station in Hunan Province.
Results
Sociodemographic profile
Land in Wangcheng County is distributed to farmers based on their family size. The
average household size is 2.5 persons, with each having an average of 0.25 ha of
cultivatable land, 90% of which is planted to rice. Among the household heads, 59%
attended primary school, 29% middle school, 8% high school, and 1% technical secondary
school; 4% were illiterate.
These farmers commonly plant two rice crops per year. Yields average 6.6 t ha -1
crop -1 , about 15% more than the provincial average. The fieldwork is mainly done by
30–40-year-old men (35%), followed by those under 30 years old (25%), and then
those 41-50 years old (20%). Many women are also actively involved in farming,
with 64% of those surveyed performing supporting labor and 5% doing all the fieldwork.
Nineteen percent do not do any fieldwork.
Knowledge on pest management
Pests and natural enemies. Farmers considered sheath blight, caused by Rhizoctonia
solani, to be their most important disease, followed by rice blast, caused by Pyricularia
oryzae, and bacterial leaf blight, caused by Xanthomonas campestris pv. oryzae. Among
insect pests, rice planthoppers—mainly brown planthopper Nilaparvata lugens and
whitebacked planthopper Sogatella furcifera— were reported as the major pest, followed
by striped stem borer Chilo suppressalis and rice leaffolder Cnaphalocrocis
medinalis. The most important rice weed cited was barnyard grass Echinochloa crusgalli
(Tables 1 and 2).
More than 90% of the farmers interviewed knew rice sheath blight, rice
planthoppers, and rice leaffolder, whereas only 5% could identify bacterial leaf streak
and 29% could identify brown spot. These findings imply that more farmers know
about insect pests than diseases (Table 2).
When asked about natural enemies, farmers most often cited frogs (93%) and
spiders (38%). Thirty-two percent of them also specified birds as being beneficial
creatures. Only one farmer knew that the lady beetle Micraspis discolor is a predator,
with some even saying it is a pest because it eats rice pollen. Eight percent of the
farmers knew the hymenopterous parasitoids. Nearly three-fourths believed that beneficial
insects can control pests.
Pest management practices of rice farmers in Hunan, China 55

Table 1. Farmers' perceptions of the seriousness of the major pests they recognized. Wangcheng
County, Hunan Province, China, 1991 (n=266).
Pest
Very serious Serious Not so serious
Total
No. % No. % No. % (%)
Diseases
Sheath blight 251 94.4 8 3.0 4 1.5 98.9
Rice blast 8 3.0 106 39.9 8 3.0 45.9
Bacterial blight 4 1.5 29 10.9 38 14.3 26.7
False smut 1 0.4
0.4
Bacterial leaf streak 7 2.6 2 0.8 3.4
Akagere disease 4 1.5 1 0.4 1.9
Brown spot 3 1.1
1.1
Sclerotium blight
1 0.4
0.4
Downy mildew 1 0.4 1 0.4 0.8
Seedling blight 1 0.4 11 4.1 4.5
Insects
Planthoppers 188 70.7 47 17.7 27 10.2 98.5
Striped stem borer 35 13.2 101 38.0 62 23.3 74.4
Rice leaffolder 35 13.2 75 28.2 93 35.0 76.3
Thrips 8 3.0 1 0.4 4 1.5 4.9
Leafhopper 20 7.5 5 1.9 9.4
Rice bug 1 0.4 0.4
Weeds
Barnyard grass
Ground chestnut
Blurush
Knotgrass
Monochoria
Water clover
Sedge
Pondweed
109 41.0 51 19.2 15 5.6 65.8
49 18.4 25 9.4 9 3.4 31.2
31 11.7 41 15.4 17 6.4 33.5
19 7.1 28 10.5 13 4.9 22.6
15 5.6 23 8.7 20 7.5 21.8
11 4.1 16 6.0 20 7.5 17.7
5 1.9 3 1.1 4 1.5 4.5
2 0.8 3 1.1 7 2.6 4.5
Table 2. Farmers recognizing pests and natural enemies, Wangcheng
County, Hunan Province, China, 1991 (n=266).
Target
(%) SD in villages (±)
Diseases
Sheath blight 97.0 3.8
Bacterial blight 56.4 28.8
Rice blast 56.0 34.8
Bacterial stripe 4.5 10.7
Brown rot 29.1 21.1
Insect pests
Planthoppers 98.1 3.2
Rice leaffolder 91.0 11.7
Striped stem borer 79.7 18.7
Rice thrips 67.7 21.1
Natural enemies
Frogs 93.2 36.1
Spiders 37.6 32.4
Dragonflies 7.9 17.6
Parasitoids 7.5 10.3
Ladybirds 0.4 0.7
56 Shao-Shi et al

Table 3. Percentage of farmers reporting difficulties in making decisions about rice pest control,
Wangcheng County, Hunan Province, China, 1991 (n=266).
Farmers
Implementing
Township interviewed Timing of economic Recognizing
(no.) spraying threshold symptoms
levels
Chengguanzhen 28 100.0 57.1 75.0
Gaotang 26 96.2 88.5 92.3
Pingtang 28 46.4 0.0 14.3
Lianhua 24 62.5 4.2 12.5
Bairuo 41 39.0 73.2 56.1
Tongguan 25 60.0 0.0 4.0
Jinggang 47 55.3 10.6 8.5
Xianing A 25 60.0 88.0 56.0
Xianing B 22 86.4 27.3 13.6
Average 64.7 38.7 36.5
Targeting Choosing Selecting No
Township pests methods pesticides opinion
Chengguanzhen 92.9 14.3 10.7 0.0
Gaotang 30.8 53.9 61.5 0.0
Pingtang 14.3 0.0 14.3 17.9
Lianhua 45.8 8.3 12.5 4.2
Bairuo 41.5 19.5 56.1 0.0
Tongguan 24.0 4.0 4.0 4.0
Jinggang 12.8 2.1 8.5 19.2
Xianing A 12.0 0.0 56.0 8.0
Xianing B 4.6 4.6 13.6 9.1
Average 30.8 11.7 33.8 7.5
Chemical control. Some 69% of the farmers for the first crop and 97% for the
second crop believed that pests can cause serious yield losses without chemical control.
Nearly all (94%) thought pests can only be controlled by using pesticides.
Farmers were also asked about their views on not using pesticides during the first
30 days after transplanting. Most (66% for the first crop and 94% for the second crop)
thought it was impossible to obtain high yields without applying pesticides during
this period.
Farmers’ difficulties in decision making for pest control were also investigated.
Farmers considered determining a spraying schedule, recognizing symptoms caused
by targeted pests, and implementing economic threshold levels as areas in which they
had difficulties. Many believed that determining the targeted pests for control and
selecting pesticides were the most difficult decisions to make (Table 3).
Pest control practices and attitudes
In 1990, farmers applied pesticides an average of 10 times per season: 3.8 times in
seedbeds to control thrips Baliothrips biformis, striped stem borer, and diseases, and
6.2 times in ricefields to control rice planthoppers and sheath blight. Twenty-two
percent of the farmers reported having been lightly poisoned or becoming ill after
exposure to pesticides.
Pest management practices of rice farmers in Hunan, China 57

During the heading stage of the second crop in 1990, farmers reported applying
isoprocarb (70%), methamidophos (40%), MTMC (22%), “Jinggangmycin” (a locally
made product)(20%), Shachong Shuang (chemical name: (1,3)-bis (sodium thiosulfate)-Z-dimethylaminopropane)(9%),
and iprobenfos (4%). Isoprocarb and
methamidophos were being used to control insects and Jinggangmycin to control
sheath blight. A few farmers were still using illegal chlordimeform and carbofuran.
Some of the applications were a mixture of different pesticides. Farmers bought most
(92%) of the pesticides from local state-run agroproduct supply corporations.
When mixing pesticides, 57% of the farmers measured the chemicals using bottle
caps, 21% with measuring cups, and 22% estimated the amounts without using any
measuring tool.
Rat control was primarily done after transplanting the second crop, with about
one-third of the farmers using poison bait and half relying on their cats. About 20%
reported doing nothing.
As labor is still plentiful in the area, 97% of the farmers reported using hand
weeding to control weeds, but only 3% used herbicides.
When selecting pesticides, 42% relied on their own experiences and 39% on the
recommendations from the local Agro-technical Extension Station (LATES). Five
percent of the farmers said they read the pesticide bottle labels for reference, and 2%
said they relied on recommendations from pesticide dealers. Some used the advice of
their relatives and friends (9%).
Sources of information for managing pests
Farmers obtained information on pest control from different sources (Fig. 2). The
county agricultural technical extension system, which includes a plant protection unit
and a pest forecasting station, forecasts pest problems and makes pest control recommendations
to farmers. Most of the farmers (61%) got their information from township
extension personnel through visits to their homes. village meetings, and radio
broadcasts. Twenty-eight percent of the farmers observed their neighbors to obtain
information.
After receiving information about pest control, farmers responded in different
ways: 42% acted on the extension system’s recommendation, 30% based on their
own experiences, and 21% after observing the actions of their neighbors. Only 2%
used economic threshold levels as a guide, and 4% relied on calendar spraying. Before
taking control actions, 62% of the farmers reported going into the fields themselves
to look, 29% observed on the edge of the field, and 4% asked other people to
observe for them. Five percent of the farmers said they did nothing after inspecting
their fields (Fig. 2).
Farmer training
Forty-one percent of the farmers had never attended the training course on pest management
organized by the extension system; 16% had been given on-farm guidance.
More than 70% of those attending the training had passed on the knowledge they
gained to their neighbors and family members. In terms of education methods, 58%
58 Shao-Shi et al

Fig. 2. Farmers’ use of information sources for making decisions on pest control
(n=266).
preferred watching demonstrations, 19% opted for attending training courses, and
14% expressed interest in studying technical materials on their own.
Discussion
The economy of Wangcheng County mainly relies on agriculture, particularly rice
production. Farmers have been making the transition from traditional to modem agriculture
during the past few years. Improving farmers’ knowledge and skills is critical
for further economic development.
Pest management practices of rice farmers in Hunan, China 59

Fu (1994) found that of 17 new techniques presented to farmers in Hunan Province,
an average of 21% of the farm households adopted them. Nearly all (95%) adopted
hybrid varieties, 56% used plastic films for covering fields, but only 21% adopted
IPM. To understand the relationship between techniques and adoption rates, Fu classified
techniques as simple and complex. The mean adoption rate for simple techniques
(38%) was significantly higher than that for complex techniques (17%). The
adoption rate for simple techniques was not related to education level, while that for
complex techniques was related to education level. IPM was considered a complex
technique.
The survey results indicated that farmers lacked knowledge on natural enemies,
economic thresholds, and yield loss estimation, and that they also overestimated the
damage caused by leaf-feeding insects. Nearly three-fourths of them thought that
spraying pesticides was the only effective way to control pests, and only 2% believed
that resistant varieties could reduce pesticide applications. More than 70% sprayed
pesticides within 30 days after transplanting. These results indicate that it would be
difficult to implement IPM without first improving farmers' knowledge and attitudes
toward IPM.
In places in China other than Hunan, an agricultural technical extension system
is responsible for disseminating information to farmers. The system has used traditional
approaches for getting information to farmers, such as village meetings, radio
broadcasts, notice boards, and demonstration plots. Although 61 % of the farmers obtained
information from these extension sources, 42% of them made decisions about
pest control based on these and 40% used pesticides based on the extension recommendations.
Most farmers preferred to make decisions based on their own experiences
or those of their neighbors. This pattern shows farmers’ low capacity for risk
taking and their relatively fixed mindset to do things based on their experiences.
The traditional top-down approach for disseminating information needs to be
changed if farmers are to expand their knowledge and adopt IPM practices. Farmer
field schools and farmer participatory research consider farmers to be the core for
implementing IPM and improving their knowledge through field practices and participatory
experiments (Escalada and Heong 1993). These methods provide good examples
of techniques to complement methods being used by extension.
References
Du Z. 1991, Tactics for integrated management of pests and diseases for rice in China. Zhejiang
Science and Technology Press, China. (In Chinese.)
Escalada MM, Heong KL. 1993. Communication and implementation of change in crop protection.
In: Crop protection and sustainable agriculture. Ciba Foundation Symposium 177.
Chichester: J. Wiley & Sons. p 191-207.
Fu T. 1994. Analysis on situation and impact of farmers adopting agriculture. New technology
in Hunan. In: Diangxian F, editor. Research on agro-tech-extension. China: China Agro.
Sci. Press. p 168-174. (In Chinese.)
Litsinger JA, Price EC, Herrera RT. 1980. Small farmer pest control practices for rainfed rice,
corn, and grain legumes in three Philippine provinces. Philipp. Entomol. 5(1-2):65-86.
60 Shao-Shi et al

Local Records Compile Committee of Hunan. 1991. Hunan yearbook. China: Xinhua Press of
Hunan. p 499-500. (In Chinese.)
Heong KL. 1984. Pest control practices of rice farmers in Tanjung Karang, Malaysia. Insect
Sci. Applic. 5:221-226.
Xu W. 1989. Theory of agricultural extension. China: Beijing Agric. Univ. Press. (In Chinese.)
Zhu Xigang. 1992. An analysis of restrictive factors of agro-technology system in China. Manage.
Agric. Sci. Technol. 5:17-21. (In Chinese.)
Notes
Authors’ addresses: Li Shao-Shi, Hunan Plant Protection and Quarantine Station, Changsha
410005, China; Guo Yu-Jie, Biological Control Institute, Chinese Academy of Agricultural
Sciences, Beijing 10081, China; Hu Guo-Wen, China National Rice Research Institute,
Hangzhou 310006; Liang Di-Yun, General Station of Plant Protection. Ministry of
Agriculture, Beijing 100026, China.
Acknowledgments: The authors thank Drs. K.L. Heong and M.M. Escalada for their support for
this project, Prof. Cheng Jiaan for his helpful comments, and the seven agrotechnicians of
the Plant Protection Center of Wangcheng County who conducted the survey. The Swiss
Agency for Development and Cooperation funded this research through the Rice IPM
Network.
Citation: Heong KL, Escalada MM, editors. 1997. Pest management of rice farmers in Asia.
Manila (Philippines): International Rice Research Institute.
Pest management practices of rice farmers un Hunan, China 61

CHAPTER 4
The role of women
in rice pest management
in Zhejiang, China
Hu Ruifa, Cheng Jiaan, Dong Shouzhen, and Sun Yinyin
Results of a farmer survey carried out in Zhejiang Province, China, in
1996 showed that as the rural economy developed, more women
became involved in rice pest management. But economic and social
factors—particularly income, household size, amount of farmland,
men’s and women’s jobs, and age—affected the proportion of women
taking responsibility for pest management. The technical service
system (the extension system) was the main source of information
women used to make decisions about pest management. Although
most of the women believed they could make pest management
decisions based on their own experience and through the information
provided by the technical service system, pesticides were overused
mainly because of inaccurate perceptions of the relationship
between control effectiveness and pesticide use.
Introduction
Women’s role in producing rice has changed dramatically since the founding of the
new China in 1949. In the old China, most farmers did not have their own land, and
they had to work for landlords as hired laborers or tenants. Women were worse off
than men. They were looked down on both politically and economically (Li et al
1985). If farmers had their own land to cultivate, the heads of the households—normally
men—made the decisions about agricultural production; women had no say in
these matters (Fei 1939).
In the early 1950s, a land reform campaign was launched in which all farmers
obtained their own land to work on. At the same time, the government adopted measures
to ensure equal rights for both men and women. The division of labor within
each family, however, varied (Li et al 1985).
From the late 1950s to the end of the 1970s, farmland became collective property
in an effort to increase production. As a result, commune team leaders—rather than
the individual farmers—made all the decisions about rice production (Lin 1990).
63

CHAPTER 2
Pest management practices
of lowland rice farmers in Cambodia
G.C. Jahn, P. Sophea, K. Bunnarith, and P. Chanthy
Introduction
We interviewed 1,265 lowland rice farmers in Cambodia on their
pest management practices from June 1995 to April 1996. Fortythree
percent of the farmers were not aware of natural enemies of
rice pests; 59% thought that pesticides increased rice yields. Only
19% believed the insecticide applications could produce pest outbreaks
by killing natural enemies. More farmers applied pesticides
in the dry season than in the wet season, and more men than women
used pesticides. Pesticide users and nonusers did not differ significantly
in average age, education, or farming experience. Nationally,
22% of wet-season farmers and 57% of dry-season farmers used
pesticides. The rice yields of pesticide users and nonusers did not
differ significantly, except in the wet season among farmers who did
not use fertilizers. Insecticides and rodenticides were the most commonly
used pesticides. None of the farmers used fungicides. An
estimated 224,000 liters of pesticides were used annually on rice.
In both the wet and dry seasons, nearly half of those applying pesticides
used a knapsack sprayer. The average reported dry-season
rice yield of 2.5 t ha -1 was significantly higher than the average reported
wet-season yield of 1.4 t ha -1 .
Pest management in Cambodia
Several surveys have been conducted in Cambodia in recent years to address aspects
of determining farmers’ knowledge, attitudes, and practices concerning rice pests.
The first of these, made by the International Rice Research Institute in 1989 (Rapusas
et al 1989), revealed that 28% of the 42 wet-season (WS) rice farmers interviewed in
Takeo Province used insecticides. This survey, however, was conducted in areas where
the government distributed insecticides to farmers, so the sample may have been biased
in favor of insecticide use. Prom (1993) found that 40% of 60 dry-season (DS)
rice farmers surveyed in Kandal Province used insecticides and that 70% believed
insecticides increase yields. Joshi et al ( 1994) reported that of the nearly 100 WS and
35

tection specialists. When a pest infestation occurs, the technicians provide timely
information about pest management to farmers (ZJNCTJNJ 1991).
Methods
To get a real picture of the role of women in rice pest management, a survey questionnaire
was designed to obtain information on farmers’ characteristics, rice production,
and pest management. The aim was to understand the knowledge, attitudes, and practices
of women in various regions within the province under different economic conditions.
A three-stage sampling method (province, county, and village) was used. Fifteen
counties were selected. From each county, one or two villages were selected and
from each village 12–30 farmers were chosen randomly. The total sample consisted
of 430 farmers.
The sample counties were chosen from more developed regions in the province,
such as Yuoqing and Dongyang counties, less developed regions, including Yongkang
and Yuyao counties, and poor areas, such as Jinyun and Panan counties. Basic information
on the survey villages in each county was gathered (Table 1).
During the survey, both women and their husbands were interviewed to determine
who was the main decision maker on each point related to pest management.
Table 1.
Basic information for site investigated based on each county surveyed. a
Farmers’
family
Land area Farming Education
County Sample County Number for every Annual income to Age of of
size type person income total head head
(ha) (10,000 yuan) b income (yr)
(%)
Jinyun 29 Mountain 4.3 0.048 0.74 77.1 51.6 4.3
Quixian 30 Hilly land 4.3 0.066 1.16 36.9 45.3 5.4
Longyou 27 Hilly land 4.2 0.054 1.34 76.9 42.0 6.6
Yongjia 18 Mountain 5.8 0.031 1.39 32.0 49.4 4.3
Anji 29 Plain 3.6 0.113 1.40 49.7 40.7 4.1
Panan 25 Mountain 3.8 0.038 1.72 59.4 45.8 5.2
Xiaoshan 60 Near city 4.2 0.041 1.75 17.3 47.0 5.7
Jiaxing 53 Near city 4.2 0.113 1.87 52.1 43.5 6.9
Yongkang 31 Plain 3.8 0.039 1.87 26.4 46.6 5.9
Yuyao 28 Plain 4.4 0.057 1.89 57.6 45.3 5.5
Yiwu 26 Plain 4.4 0.035 2.75 14.3 45.4 6.6
Zhuji 12 Plain 3.6 0.036 2.92 22.3 43.6 8.8
Ouhal 25 Near city 4.6 0.041 2.96 18.2 43.0 8.0
Dongyang 18 Plain 3.7 0.029 4.51 37.8 45.8 6.5
Yuoqing 19 Plain 4.9 0.034 5.43 18.5 48.9 4.4
a Numbers in table are average values for farmers surveyed in the sample villages except for sample size.
b US$1 = 8.30 yuan in 1996.
Source: Survey.
The role of women in rice pest management in Zhejiang, China 65

The survey data were encoded and analyzed using the spreadsheet program Microsoft
Excel ® .
Results
Role of women in pest management
The results of the survey indicated that men were still the main decision makers on
matters related to applying pesticides. Only 17% of the households had women making
decisions about pest management, whereas 6% of the households reported that
both men and women made the decisions together. Only 17% of the households had
women taking responsibility for applying pesticides, but 46% of the families reported
that women made decisions on pest management.
Women's knowledge, attitudes, and practices
Major pests. The women farmers interviewed reported that rice sheath blight (56.9%),
planthoppers (42.6%), and Echinochloa crus-galli (58.0%) were the most severe pests;
rice blast (31%), stem borer (29.5%), and Cyperaceae eleocharis (8%) were the second
most severe; and rice bacterial blight (5.2%), leaffolder (19.7%), and Marsilea
quadrifolia (8%) followed.
Main control methods. Most women farmers preferred chemical control. Some
93.2% of women farmers mentioned chemical control as their first control method for
diseases, 87.5% used it for insect pests, and 81.8% applied it for weed control. The
next preferred control methods were using resistant varieties for diseases (5.7%) and
insect pests (19.3%), and manual weeding for weed control (14.8%).
Number of pesticide applications. Women farmers averaged 2.75 pesticide sprays
for the first rice crop season and 3.69 for the second.
Estimation of amount of pesticides to be used. Among women farmers who decided
the amount of pesticides to be used based on their experience, 62.5% measured
them with special tools, such as measuring cups, whereas 35.2% simply estimated the
amount. Some 15.9% would add more pesticides, because they thought that more
pesticides would provide more effectiveness.
Cocktail pesticides. Farmers usually preferred cocktail pesticides. Some 73.8%
thought it could save labor and time by controlling several pests together, whereas
14.8% believed that cocktail pesticides were more effective. More than 61% of the
women farmers mixed pesticides to make a cocktail—17.1% combined three, and
11.4% used more than three compounds.
Attitude for early pesticide application to control leaf-feeding insects. Almost
three-fourths (73.9%) of the women farmers interviewed thought that damage caused
by leaf-feeding insects at the early stage of the rice crop could result in severe yield
losses and 94.6% of them believed that applying insecticides to control leaf-feeding
insects at the tillering stage was necessary.
66 Ruifa et al

Table 2. Information source for decision making in rice pest management.
Question
Selfobservation
or experience
Village
leaders or
technicians
Broadcasting
Blackboard
Neighbors
What pest to 46.1
control
When to control 20.2
How many times 24.7
to control
Which pesticides 3.4
to use
How much pesticide 5.6
to use
Source: Survey.
41.8 55.0
23.6 14.0
29.1 27.0
2.7 1.0
2.7 3.0
54.6 58.6
16.2 17.2
27.3 21.2
2.0 1.0
0 2.0
Information source for pest management
Table 2 shows the information sources of women farmers for pest management. More
than 60% of the women obtained advice on pest management from technology service
organizations, which include village leaders or technicians (29.3%) and pest
control services (34.3%). Some 29.3% of the women got advice from neighbors or
relatives. Only 5% of the women obtained the information from pesticide sales agents
and 4% sought advice from their relatives.
A comparison of the main information sources for decision making on key issues
in pest management, such as target pests, timing and number of applications, and type
and amount of pesticides to be used, showed that more than half of the women farmers
thought that they could make decisions on the number of applications and type
and amount of pesticides to be used based on their own experience. On the other
hand, more than half of the women farmers indicated that they obtained information
from the technical service system for decision making on timing of applications. Radio
broadcasts were the most important information source, followed by village leaders
and technicians.
Decision-making behavior
More than half (59.6%) of the women farmers indicated that they would decide to
control pests if they perceived that pests would infest their crop, whereas 30.8% of
them said they would make a control decision only when pest infestation actually
occurred in their own field or in neighbors’ fields. This implies that their decisions to
control pests were influenced by seeing pests (Table 3).
The main reason given by women farmers for controlling pests was to avoid
yield losses (Table 3). Other farmers said that they would decide not to carry out
control actions if they perceived the loss to be small and if they had no time. All three
reasons are closely related to economic benefits, directly through yields or indirectly
through savings in time and labor, which implies that economic benefits drive farmers’
pest management decisions.
The role of women in rice pest management in Zhejiang, China 67

Table 3. Women's decision-making behavior in rice pest
management.
Frequency
of making
decision (%)
When decision was made
Before planting 9.6
Before pest occurs in own field 59.6
After pest occurred in neighbors' field 13.8
After pest occurred in own field 17.0
Reason for control decision
To avoid yield reduction
Following others
Cadre order
85.4
9.0
5.6
Reason to not make control decision
Too busy to control pest 44.8
Smaller loss for pest 46.3
Control cost is higher than loss 4.5
Pesticide harms health 4.5
Varieties have enough resistance 3.0
Source: Survey.
Factors affecting women’s role in rice pest management
As the rural economy develops, more and more women take part in rice production
and pest management activities (Lei et al 1990, Li et al 1985). Moreover, socioeconomic
factors influence their involvement in rice pest management.
Economic factors. Farmers’ economic conditions affect the role of women in
decision making on pest management. The data were sorted by levels of average income
per household based on total interviewed households and the average income
per household per county. Table 4 shows that the distribution of women making decisions
increased at first, then declined with increases in income. The highest proportion
of women farmers who made decisions was at an income level of $1,800-2,400
per household. The same trend was found in other counties. The highest percentage of
women who made decisions was in Yongkang, where average income per family was
$2,240.
In general, the proportion of farming income to total income declined as household
income increased (Table 4), which indicates that other family members, usually
men, could derive income from nonagricultural industries. In such a situation, women
have to assume responsibility for rice production, which includes making decisions
on production and pest management. When pests are observed, women believe that it
is necessary to control them.
As rural economies develop, both men and women have more opportunities to
work in local industries and their main income is derived from nonagricultural industries.
As a consequence, they transfer their lands to other farmers or hired workers to
do field work. In high-income families, the main decision makers are men, because
68 Ruifa et al

Table 4. Changes in role of women in rice pest management.
Pest management
Household
Farmers annual
surveyed income
(no.) (10,000
yuan) a
Farming
income/
total
Income
(%)
Decision
making
(%) b
Pesticide spraying
work
Done by Done by
women employee
(%) c (%)
By household income level (10,000 yuan d )
0-0.5 36
0.6-1.0 122
1.1-1.5 96
1.6-2.0 70
2.1-3.0 62
3.1-5.0 26
57.4
49.9
42.1
37.0
34.2
26.1
5.6
13.1
19.8
27.1
16.1
0
21.6 2.7
14.8 0
7.3 3.1
20.0 10.0
29.0 4.8
20.0 12.0
By county
Jinyun 29 0.74 77.1 20.7 13.8 0
Yongria 18 1.39 32.0 27.8 50.0 0
Yongkang 31 1.87 26.4 35.5 38.7 6.5
Ouhai 25 2.96 18.2 12.0 44.0 32.0
Dongyang 18 4.51 37.8 16.7 5.6 16.7
Yuoqing 19 5.43 18.5 10.5 5.3 15.8
a Number is average data among farmers surveyed in the county.
b Includes decisions made by men and women
both. c Means work done by men and women by themselves. The amount of work that is done by both men and
women is not included. d US$1 = 8.30 yuan in 1996.
Source: Survey.
men have more land or women hold nonfarming jobs. The proportion of families in
which women are the major decision makers tends to decline with income.
Table 4 shows that when the proportion of farming income to total income was
higher (over or nearly half) because of lower household income (

Fig. 1. Impact of social factors on women’s frequency in decision making for rice
management: (A) home size, (B) farm size, (B) women’s age, (D) job type ( £ men’s, women’s).
who made decisions decreased. In families of 2-3 people, 24.8% of the women planned
rice production and made pest management decisions. In families of 5-6 people, the
proportion decreased to 20.6%.
Farm size. Farm size reflects the farm area cultivated by each family. In Zhejiang
Province, most farmers are not full-time field workers because of the small farm size—
0.125-0.25 ha per family on average. Some farmers with steady jobs in other industries
transferred their farmland to other farmers, thus increasing the landholding of
some farmers. Such farmers derived their incomes mainly from rice production. In
these households, the men were the main workers in fields and women had fewer
opportunities to make decisions. Figure 1B shows that the proportion of women who
made decisions in pest management decreased as farm size increased. When the average
farmland per family is less than 0.2 ha, 32.6% of the women made decisions on
rice pest management. When farmland per family is above 0.6 ha, the proportion
decreased to 11.6%.
Age. The role of women in pest management decision making was also affected
by their age. Old women are often traditional women who have always had a lower
position in agricultural production (Wang 1996, Zhou 1996). Because their husbands
or others mainly produce rice, these women make fewer production decisions. In
contrast, younger women tend to be able to produce rice relying on their own knowledge
and they make many production decisions. Fig. 1C shows that the proportion of
women who made decisions in rice pest management decreased (from 17.0% to 12.3%)
as their age increased.
70 Ruifa et al

Fig. 1. Impact of social factors on women’s frequency in decision making for rice
management: (A) home size, (B) farm size, (B) women’s age, (D) job type ( £ men’s,
women’s).
who made decisions decreased. In families of 2–3 people, 24.8% of the women planned
rice production and made pest management decisions. In families of 5-6 people, the
proportion decreased to 20.6%.
Farm size. Farm size reflects the farm area cultivated by each family. In Zhejiang
Province, most farmers are not full-time field workers because of the small farm size—
0.125–0.25 ha per family on average. Some farmers with steady jobs in other industries
transferred their farmland to other farmers, thus increasing the landholding of
some farmers. Such farmers derived their incomes mainly from rice production. In
these households, the men were the main workers in fields and women had fewer
opportunities to make decisions. Figure 1B shows that the proportion of women who
made decisions in pest management decreased as farm size increased. When the average
farmland per family is less than 0.2 ha, 32.6% of the women made decisions on
rice pest management. When farmland per family is above 0.6 ha, the proportion
decreased to 11.6%.
Age. The role of women in pest management decision making was also affected
by their age. Old women are often traditional women who have always had a lower
position in agricultural production (Wang 1996, Zhou 1996). Because their husbands
or others mainly produce rice, these women make fewer production decisions. In
contrast, younger women tend to be able to produce rice relying on their own knowledge
and they make many production decisions. Fig. 1C shows that the proportion of
women who made decisions in rice pest management decreased (from 17.0% to 12.3%)
as their age increased.
70 Ruifa et al

The survey likewise showed that farmers evidently overused pesticides. Farmers
preferred to use pesticide cocktails, mixtures of insecticides and fungicides, to save
time and labor for applications. The dosages of pesticides used were also very high
because farmers misunderstood the relationship between recommended dosage and
effectiveness, and lacked knowledge on measuring pesticides. An interesting finding
was that the rationale for overusing pesticides was the knowledge gap among women
farmers, who felt confident in their pest management decisions based on their observations
and experience. This implies that there were knowledge gaps between what
farmers knew and should know about integrated pest management (IPM).
One knowledge gap is the lack of an ecological understanding of pest problems.
When farmers made decisions on control, they first considered economic benefits
because they wanted to avoid yield losses. They did not take any control actions when
they perceived only a small crop loss and they lacked time to carry out pest control
because they were busy with their other nonagricultural jobs. When farmers made
decisions on the type of pesticide to apply, they first considered saving time and labor
by mixing several insecticides and fungicides, because they thought that this would
be beneficial. But they did not consider ecological conditions as essential in deciding
whether or not to take a control action or to use a pesticide mixture.
Another gap is information dissemination between the extension system and farmers,
or between research and extension. Although the extension system is the main
information source for farmers, only half of them said they obtained information from
the extension system when they planned to make a pest management decision. But
the decisions farmers made were different from the recommendations made by the
IPM program for two possible reasons: farmers did not use the recommendations
from the extension system or the extension system did not make correct recommendations
according to the IPM program.
Two related areas could therefore be explored further: (1) improving farmers’
knowledge, especially their ecological understanding of pest problems, and (2) understanding
the information flow between research, extension, and farmers, to find
out where the gaps are and ways to improve information dissemination. The results
from this further study will provide us with more detailed information that can be
used to refine an IPM program.
References
Ellis F. 1988. Peasant economics: farm households and agrarian development. Cambridge (Great
Britain): Cambridge University Press. 257 p.
Epstein CF. 1971. Women’s place. Berkeley (Calif., USA): University of California Press.
221 p.
Evenson RE. 1985. Observations on institutions, infrastructure technology and women in rice
farming. In: Women in rice farming. Great Britain: Gower Publishing Company. p 21-36.
Fei X. 1939. Jiangcun economics: peasant life in China. Nanjing (China): Jiangsu People’s
Press. 385 p.
72 Ruifa et al

Kada R, Kada Y. 1985. The changing role of women in Japanese agriculture: the impact of new
rice technology on women’s employment. In: Women in rice farming. Great Britain: Gower
Publishing Company. p 37-54.
Lei J, Yang S. Cai, W. 1990. The change of the marriage and the family in the Chinese countryside
since the reform of the economics system. Beijing (China): Peking University Press.
502 p.
Li C, Feng R, Xu X. 1985. ‘Half-sky’ role of China’s women in rice farming systems. In:
Women in rice farming. Great Britain: Gower Publishing Company. p 65-70.
Lin JY. 1990. Prohibition of factor market exchanges and technological choice in Chinese
agriculture. IRRI Working Paper. Los Baños (Philippines): International Rice Research
Institute. 27 p.
Pradhan B. 1985. The role of women in household production systems and rice farming in
Nepal. In: Women in rice farming. Great Britain: Gower Publishing Company. p 257-286.
Res L. 1985. Changing labor allocation patterns of women in rice farm households: a rainfed
rice village, lloilo Province, Philippines. In: Women in rice farming. Great Britain: Gower
Publishing Company. p 91-118.
Sajogyo P. 1985. The impact of new farming technology on women’s employment. In: Women
in rice farming. Great Britain: Gower Publishing Company. p 149-170.
Unnevehr LJ, Stanford ML. 1985. Technology and the demand for women’s labor in Asian rice
farming. In: Women in rice farming. Great Britain: Gower Publishing Company. p 1-20.
Wang J. 1996. The gender difference for household’s position of rural women in Zhejiang.
China Zhejiang Acad. J. 1996(6).
Wang X, Zheng K. 1993. Why farmers leave uncultivated land: thinking of rural reform and
development in the 1990s. Taiyuan (China): Shanxi Economics Press. p 375-384.
Watson GA. 1985. Women’s role in the improvement of rice farming systems in coastal swamplands.
In: Women in rice farming. Great Britain: Gower Publishing Company. p 187-208.
White B. 1985. Women and the modernization of rice agriculture: some general issues and a
Javanese case study. In: Women in rice farming. Great Britain: Gower Publishing Company.
p 119-148.
Xiong Z, Cai H. 1992. Rice in China. Beijing: China Agricultural Press. 606 p.
Zhou, Y. 1996. The generation cultural characteristic of rural society in China today. Zhejiang
Acad. J. 1996(2).
ZJNCTJNJ (Zhejiang Nongcun Tongji Nianjian) (Zhejiang Rural Statistical Year Book). 1991-
1995. Zhejiang Statistical Bureau 1991 -1995. Hangzhou (China): ZJNCTJNJ.
Notes
Authors’address: Zhejiang Agricultural University, 310029 Hangzhou, People’s Republic of
China.
Citation: Heong KL, Escalada MM. editors. 1997. Pest management of rice farmers in Asia.
Manila (Philippines): International Rice Research Institute.
The role of women in rice pest management in Zhejiang, China 73

CHAPTER 5
Pest management practices
of rice farmers in Tamil Nadu, India
S.D. Sivakumar, S.R. Subramanian, S. Suresh, and M. Gopalan
We conducted a survey of 120 farmers in rice-growing areas in
Thanjavur and Chengai MGR districts in Tamil Nadu to study their
rice pest monitoring, assessment, and control practices. We also
determined the constraints to their adoption of integrated pest management
(IPM) practices. Leaffolders, stem borers, and rice bugs
were commonly observed rice insect pests, and dragonflies and damselflies
were commonly mentioned beneficial insects. Department
of Agriculture officials were the major source of pest control information.
All the farmers used various nonchemical IPM measures,
including insect-resistant varieties, cleaning bunds, and synchronized
planting. The cost of chemical control was higher on small farms
than on large farms. A lack of easy-to-follow pest assessment techniques
and lack of community efforts were cited as the major constraints
to following IPM practices.
Introduction
Rice is the staple food in Tamil Nadu, India. Rice was planted on 2.6 million ha in the
state in 1970-71 (35% of the total area cropped) but on only 2.3 million ha in 1993-
94. Rice production, however, increased from 5.0 to 6.6 million t during this period,
primarily because productivity increased from 1.9 to 2.8 t ha -1 . This achievement has
been attributed to the significant developments in agricultural research and education,
extension, and infrastructure. Farmers’ adoption of high-yielding varieties and other
improved technologies as well as their use of chemical fertilizers have undoubtedly
contributed to increased rice production; pest problems, however, have also accompanied
these advances.
With the widespread use of synthetic organic pesticides, pests prevalent in traditional
tall indica rice cultivars, such as the swarming caterpillar Spodoptera mauritica
Bdv., grasshopper Hieroglyphus banian F., and spiny beetle Dicladispa armigera,
became noticeably absent in fields planted to modem semidwarf cultivars. Instead,
75

minor pests, such as brown planthopper Nilaparvata lugens Stäl. in the kuruvai season
(June-July sowing) and rice leaffolder Cnaphalocrocis medinalis (Guen) in the
samba season (August-September sowing), became major pests. Yellow stem borer
Scirpophaga incertulas (Wlk.) became endemic in the November-February crop. Diseases
such as bacterial leaf blight appeared after cultivar TN1 was introduced during
1965. Rice tungro virus, a virus disease transmitted by the green leafhopper, was
epidemic in 1984 in Tamil Nadu.
Today, more than 30 insects and 10 noninsect pests cause damage to the rice
crop. Ramasamy et al (1993) estimated the loss in rice production in Tamil Nadu from
insect pests to be 470,000 t and from diseases 270,000 t. Among these pests, farmers
reported leaffolder to be the most serious, causing 24% of the loss, followed by earhead
bug (19%) and yellow stem borer and gall midge Orseolia oryzae (W.M.) (both 13%).
Rice blast, an endemic disease in Tamil Nadu because of high relative humidity, reportedly
causes 24% of the loss. Pests appear to be a major constraint to rice production.
Pest control concepts
Over the years, researchers have focused on developing effective pest management
measures. By 1948, DDT dust formulations were widely used for controlling green
leafhopper, brown planthopper, yellow stem borer, and gall midge. The era of organophosphorus
insecticides began in 1952 with the testing and use of methyl parathion,
followed by the introduction of other insecticides for controlling various rice pests.
The concept of prophylactic treatment was given much attention because of the high
illiteracy rate among farmers and the complexities involved in identifying pests and
determining their intensities. These regular applications of insecticide and fungicide
were popular for many years.
Hand- and machine-operated sprayers, dusters, blowers, and pesticides were sold
at subsidized rates to farmers. To achieve more thorough and quicker coverage, the
Department of Agriculture organized mass ground sprayings. From 1966 to 1972,
insecticides were applied aerially using ultra-low-volume formulations of phosphamidon
and fenitrothion on 5,000–200,000 ha of rice—in addition to conventional
methods of applying pesticides. Pesticide use in Tamil Nadu reached an all-time high
of 10,926 t in 1984-85. The inappropriate use of pesticides resulted in outbreaks of
leaffolder, brown planthopper, and rice tungro virus, the development of resistance to
insecticides, and a resurgence of leaffolder to phorate and brown planthopper to
quinalphos, methyl parathion, phorate, and synthetic pyrethroid (especially
deltamethrin). Because of these problems, the need to rationalize pesticide use and
develop an integrated approach to rice pest management assumed greater significance.
Integrated pest management (IPM) concept
The popular demand for a change in pest control technology is related to the social
cost of pesticide use (Katherine 1980). In this context, IPM becomes more relevant.
IPM stresses the rational use of a combination of pest control techniques while enhancing
the role of natural regulatory mechanisms to produce an economically and
76 Sivakumar et al

socially acceptable yield with no adverse effects on the environment (Teng and Savary
1992).
In the Tamil Nadu context, several important things happened because of the
gradual adoption of IPM. Two broad-spectrum insecticides were withdrawn from the
market and penalties put in place for those manufacturing and marketing the insecticides.
Economic threshold levels were developed for major rice pests, and ecologically
sound botanical products, as substitutes for synthetic organic pesticides, were
extensively tested. Neem oil, neem cake, mahua oil, and palmrosa oil were found to
act as powerful antifeedants, growth regulators, and ovicides for rice pests. Biocontrol
agents, such as egg parasitoid Trichogramma japonicum and T. chilonis, were found
to be effective on yellow stem borer and leaffolder, respectively, and the microbial
pesticide Bacillus thuringiensis on leaffolder. Researchers also investigated the impact
of cultural practices on the population dynamics of rice pests. As a result of these
efforts, scientists and extension workers developed and disseminated several
nonchemical control measures to the farming community.
Socioeconomic studies on pesticide use decisions revealed that plant protection
measures used by farmers were generally based on their anxiety to save the crop
(Rajagopalan 1983), expectations regarding the timing and intensity of pest attack
(Seeta 1985), and source of information about plant protection (Salvi and Gajre 1985).
Farmers’ pest identification skills and their knowledge of pest control measures greatly
influenced their pesticide use (Sagar and Pal 1986, Satapathy and Pattnaik 1986). The
socioeconomic status of farmers also affected their access to information sources and
their adoption of improved plant protection measures (Sen and Doijad 1990). The
views of political and agricultural decision makers also influenced the use of pesticides
in rice (Kenmore 1987). The major constraints encountered by farmers in applying
plant protection measures were heavy rain during the wet season and ineffective
and expensive plant protection chemicals (Veerasamy et al 1992).
In view of these factors, farmers’ pest management practices, the cost of pest
control measures, and the constraints to adopting IPM practices needed to be analyzed.
The objectives of this study were to examine farmers’ pest monitoring and
assessment methods, determine their awareness of IPM practices and the pest control
methods they follow, analyze the cost of pest control measures, and identify the constraints
to their adoption of IPM.
Survey site
In 1993-94, rice was cultivated on 2.3 million ha in Tamil Nadu, yielding 6.6 million
t of grain. Of the state’s districts, Thanjavur and Chengai MGR have the largest rice
areas and produce the most rice, which were the main reasons for including them in
this study. In 1993-94, the rice area in Thanjavur was 530,000 ha (73% of the total
cropped area), with 290,000 ha (65% of the total area) in Chengai MGR. Production
was 1.14 million t in Thanjavur and 0.88 million t in Chengai MGR. Together, the
districts accounted for 36% of the state’s rice area and 30% of its production. For this
Pest management practices of rice farmers in Tamil Nadu, India 77

survey, greater emphasis was placed on Thanjavur because more of its area is dedicated
to rice production and because of the greater importance of rice in the area.
Tamil Nadu can be divided into seven agroclimatic zones. Chengai MGR district
is in the northeastern zone and Thanjavur is in the Cauvery Delta zone. The area of
Chengai MGR is 7,857 km 2 , and in 1991 the population was 4.7 million. Thanjavur
occupies 8,280 km 2 , and in 1991 had 4.5 million people. The population density per
square kilometer was slightly higher in Chengai MGR (592 persons) than in Thanjavur
(547 persons), but more of the population lives in rural Thanjavur (77%) than in rural
Chengai MGR. The literacy rate in both districts was 66%.
About 40% of the population participated in the labor force. In Thanjavur, cultivation
laborers constitute 20% of the total worker population and agricultural laborers
45%; in Chengai MGR, the numbers are 15% and 32%, respectively. An additional
13% of the workers in both districts were involved in nonrice agricultural activities,
such as livestock, fisheries, and orchards,
In 1993-94, the total cropped area was 450,000 ha in Chengai MGR (57% of the
district's total area) and 720,000 ha in Thanjavur (88% of the total area). The irrigated
areas in both districts accounted for about 80% of the total cropped area. In contrast,
canals provide irrigation for 94% of the total irrigated area in Thanjavur.
The average annual rainfall is 1,126 mm in Thanjavur, although in 1993 it was
1,712 mm; in Chengai MGR, the average is 1,165 mm, but in 1993 it was 1,398 mm.
Most of the rain was during the northeast monsoon (October to December).
Sixty-two percent of the rice area in Chengai MGR and 83% in Thanjavur is
grown in the samba season. More than 90% of the rice area is irrigated, and nearly all
is grown with high-yielding modern varieties. The average rice yield is 3.0 t ha -1 in
Chengai MGR and 2.2 t ha -1 in Thanjavur.
Thanjavur has 13 state seed farms and Chengai MGR two. In 1993-94, farmers in
Thanjavur used 54,000 t of nitrogen fertilizer and those in Chengai MGR used 35,000
t; the use of phosphorus and potash fertilizers was similarly greater in Thanjavur than
in Chengai MGR. Sixty-six percent of the 1,314 retail fertilizer outlets in Chengai
MGR were private, as were 70% of the 2,699 in Thanjavur.
In 1993-94, farmers in Chengai MGR used 81 metric tons of powdered pesticides
and 937,000 liters of liquid pesticides, while those in Thanjavur used 382 metric
tons of powdered pesticides and 86,000 liters of liquid pesticides. Pesticides were
sold through 565 retail outlets in Chengai MGR and 1,728 in Thanjavur. About 44%
of the retail outlets in Chengai MGR and 62% of those in Thanjavur were private.
Methods
Four subdistricts in Thanjavur and one in Chengai MGR were selected for the study.
Six villages were then randomly selected from the list of villages in each subdistrict.
Two villages were selected in Thiruthuraipoondi and Orathanadu, and one each in
Kumbakonam and Papanasam. In Chengai MGR, all six villages were from the
Madurandagam subdistrict.
78 Sivakumar et al

Complete lists of all rice farmers in each village were not available. Upon entering
each village, survey staff randomly selected ten farmers.
A questionnaire was developed and pretested. Personal interviews were used to
collect the data. Recent graduates of agricultural schools conducted the field survey
after being briefed on the objectives and the questionnaire. The data were collected
during 1994-95.
Results
Farmer characteristics
Respondents were stratified into those operating small (4 ha). Many (57%) owned medium-sized farms, 26% had small farms,
and 17% had large farms. The medium-sized farms averaged 2.0 ha, the small 0.59
ha, and the large 8.7 ha. About 25% were 34 years and below, 63% aged between 35
and 50, and 12% 51 years and above.
More than one quarter (27%) had an elementary education, 22% went through
middle school, and 24% reached high school. Fifty-two percent had 11–20 years of
farming experience, 16% had 25–30 years, 3% had 30 years, and 29% had 10 or less
years’ experience. Several researchers (Gogoi and Gogoi 1989, Grieshop et al 1988,
Athimuthu 1990, Shanta 1992) have reported that farmers’ ages, education levels,
years of experience, and farm sizes influence their adoption of agricultural technologies.
Of the total cropped area, rice was grown on 80% of it, chickpea on 9%, sesame
on 4%, sugarcane on 5%, and other crops on the remainder.
Rice cultivation
Rice is cultivated in Tamil Nadu during the seasons of sornawari (April-August), kar
(May-September), kuruvai (June-October), samba (August-December), thaladi (November-March),
and navarai (December-April). The area planted to rice is greatest
during the samba season. During the survey, 44% of the rice area was planted in the
samba season followed by 25% in kuruvai, 23% in thaladi, and 8% in navarai.
Short-duration cultivars are generally grown in kuruvai and navarai seasons,
whereas medium- and long-duration cultivars are grown in samba and thaladi seasons.
ADT 36, TKM 9, IR50, IR66, and PY3 were the most common cultivars grown
during kuruvai, and ADT 39, ADT 38, CO 43, and CR 1009 were most popular during
samba and thaladi seasons. During the navarai season, ADT 36, ADT 39, IR20,
IR50, and CO 43 were the most popular cultivars.
On the surveyed farms, White Ponni was grown on 21% of the rice area and CR
1009 on 18% of the area. CR 1009 was predominantly planted in Thanjavur, whereas
White Ponni was the predominant cultivar in Chengai MGR. The land planted to
ADT 36, ADT 39. IR50, and ADT 38 occupied 8-11% of the rice area.
In the kuruvai season, rice yields ranged from 3 to 4 t ha -1 on 17 farms and from
4.1 to 5 t ha - 1 on 16 farms. Similarly, during samba, thaladi, and navarai seasons, rice
yields on most of the surveyed farms ranged from 3 to 5 t ha -1 . The mean yield of
Pest management practices of rice farmers in Tamil, Nadu, India 79

4.1 t ha -1 in kuruvai was significantly more by the t-test (P = .05) than those in samba
(3.6 t ha -1 ), thaladi (3.3 t ha -1 ), and navarai (3.5 t ha -1 ) seasons.
Pest-monitoring practices
Farmers commonly observed the occurrence of pests and assessed their populations
in the field. Nearly half the farmers monitored their fields once every 2 d, 23% visited
their fields daily, and 23% went to their fields once a week. In total, 95% of the
farmers did some form of pest monitoring, although none recorded pest incidence.
The most commonly observed insect pests were earhead bug, stem borer, and
leaffolder. At tillering, leaffolders were observed on 28% of the farms and stem borers
on 33% of them. Earhead bugs were seen during the milk and dough grain stages,
and jassids during the seedling, flowering, milk grain, and dough grain stages. A few
farmers observed brown planthoppers and blast disease during flowering. Additionally,
the majority of the farmers noticed dragonflies, damselflies, and spiders, which
they reported do not damage crops.
Pest management practices
Pest control activities following pest monitoring involved awareness about appropriate
control measures and adopting pest management practices. Farmers learned about
pest management practices from several sources. Department of Agriculture officials
served as the primary source for information on pest management practices, with
76% of the farmers seeking information from them. Other farmers sought pest management
advice from neighboring farmers and agricultural input dealers, and others
(7%) said they were guided by their own experiences. Mohankumar (1985),
Subrahamanyam (1985), and Goodell et al (1984) reported similar findings.
The extent of farmers’ awareness and adoption of IPM practices varied (Table 1).
Among these practices, all were using insect-resistant cultivars, cleaning bunds, and
following a synchronized planting schedule. The Department of Agriculture's efforts
to popularize the use of high-yielding cultivars have resulted in farmers adopting
them, although most could not relate the variety to the pest to which it is resistant.
Cleaning bunds is a traditional practice that farmers have generally practiced for a
long time, and canal and tank irrigation have enabled farmers to synchronize their
planting schedules.
More than 90% of the farmers knew about the practices of proper spacing, weekly
pest observation, avoiding the use of nitrogen fertilizer during pest attack, and how to
weed properly. Although transplanting was not usually done with ropes to mark proper
spacing, farmers said they made sure adequate spacing was observed.
Organic manure supplied the major source of nutrients 30 years ago, but when
high-yielding cultivars were introduced, the use of inorganic fertilizers gained momentum.
The farmers surveyed applied nitrogen fertilizer in three splits (one basal
and two topdressed) after each weeding. Only hand weeding was predominantly practiced,
with women doing most of the work.
Thirty-three percent of the farmers clipped leaf tips for old seedlings, but not
many knew the pest control reason for the practice. Few farmers were aware of the
80 Sivakumar et al

Table 1. Pest management practices of farmers (no.) a , Thanjavur
and Chengai MGR districts, Tamil Nadu, India, 1993-94.
Practice Awareness Adoption
Summer plowing
Using insect-resistant cultivars
Treating seeds
Cleaning bunds
Clipping leaf tips before transplanting
Organizing synchronized planting
79
(66)
120
(100)
86
(72)
120
(100)
55
(46)
120
(100)
120
(100)
112
47
(39)
120
(100)
58
(48)
120
(100)
39
(33)
120
(100)
Proper spacing
112
(931
Managing irrigation judiciously
74
(83)
(62)
Avoiding use of nitrogen fertilizer 110 110
during pest attack (92) (92)
Using neem-coated fertilizer 96 69
(80) (58)
Weed treatment 116 116
(97) (97)
Using light traps to monitor pests 69 3
(58) (3)
Weekly pest observation 116 116
(96.7) (96.7)
Observing predatory organisms 76 32
(63.3) (26.7)
Avoiding using pesticide when 47 13
predators are more numerous (39.2) (10.8)
Application of insecticides 73 14
based on ETL (60.8) (11.7)
Use neem oil for pest control 24 21
(20.0) (17.5)
Avold using resurgence-inducing 50 50
insecticides (41.7) (41.7)
Release of biological agents 12 8
(10.0) (6.7)
Use biomodiolone 37 20
(30.8) (16.7)
Use zinc phosphide 55 50
(45.8) (41.7)
Digging and killing rats 99 99
(82.5) (82.5)
Harvesting rice closer to soil 85 45
(70.8) (37.5)
a Numbers in parentheses indicate percentage to total sample of farmers.
Pest management practices of rice farmers in Tamil Nadu, lndia 81

practices of summer plowing, light traps, and neem-coated urea; consequently, their
adoption was low, with many perceiving them to involve additional expense.
Although more than 60% of the farmers were aware of applying insecticides
based on economic threshold levels and monitoring populations of predatory organisms,
few had actually adopted the practice. Thirty-nine percent of the farmers knew
not to apply pesticides when more predators were seen, but adoption was low because
most did not understand the concept of predation or biological control. Using neem
oil for pest control and releasing biocontrol agents were practices known to only a
few farmers, and their adoption was likewise very low.
Most of the farmers (83%) controlled rodents by digging into their burrows and
then killing them, a practice more common in Chengai MGR than in Thanjavur. Only
38% of the farmers harvested rice close to the soil even though 71% were aware of the
pest management benefits. Farmers cited labor problems during harvest as the reason
for limited adoption of this practice.
Farmers’ preferences for pesticides targeted at specific pests differed (Table 2).
Farmers mainly used BHC 10% to control earhead bug, jassids, and brown planthopper,
whereas phosphamidon was used to control stem borer. Quinalphos, methyl parathion,
and monocrotophos were used to control leaffolder, and edifenphos was used to control
rice blast.
The cost of chemical control per hectare broken down according to farm size was
US$15 for small farms, US$14 for medium farms, and US$14.40 for large farms.
Expenditure on chemical control was significantly higher on small farms than on
medium and large farms, although it was not significantly different between medium
and large farms. Mode of purchase accounted for variation in chemical costs for small,
medium, and large farms. Farmers with medium or large holdings bought pesticides
in larger quantities than did those with small farms, resulting in lower prices. The
difference in application cost mainly depended on whether farmers used powders or
sprays, because the cost of dusting involves less labor than does spraying. No farmers
applied pesticides more than twice during a season. Spraying was usually done within
a week after observing a pest problem.
About 26% of those surveyed—mostly with large farms—had their own sprayers.
Others hired professionals to apply pesticides.
Table 2. Type of pesticide applied by farmers (no.) targeted at a specific pest, Thanjavur and
Chengai MGR districts, Tamil Nadu, India, 1993-94.
Pesticide Leaf- Leaf- Stem Gall Jassid Earhead Brown Blast
folder hopper borer midge bug planthopper
Monocrotophos 9 1 12 – – 1 – Endosulfan 5 – 1 1 1 2
– –
Dichlorvos – – 1 – – – – –
Phosphamidon 3 3 28 – – 3 – –
BHC 10% (Lindane) – – – – 11 33 7 –
Quinalphos 18 – – – –
–
– –
Methyl parathion 13 – 3 – – –
– –
Edifenphos – – – – – – – 9
82 Sivakumar et al

Constraints to adoption of pest management practices
Three-fourths of the farmers expressed lack of an easy pest assessment technique as
the major constraint in following need-based application of pesticides. Additional
costs and lack of knowledge about IPM practices were the other major constraints
expressed. Most of the farmers were also not aware of the use of biocontrol agents.
Discussion
The survey farmers commonly look for pests and symptoms of their damage during
field visits. The most commonly observed pests are earhead bug, stem borer, and
leaffolder. Farmers commonly see dragonflies, damselflies, and spiders and know
that they are insects that do not damage rice.
Department of Agriculture officials are the primary source of pest management
information for farmers. Among those pest management practices adopted by farmers,
few could give the reasons behind the practices, except for cleaning bunds, proper
spacing, and proper weeding.
The adoption of biological control and some of the physical practices, such as
summer plowing, clipping leaf tips, and harvesting rice close to the soil level, is very
low. Shantha (1992) reported similar findings. Keith et al (1993) found that even
farmers who applied the same amount of pesticides with improved timing of their
applications obtained increased yield.
The farmers mostly make decisions about applying pesticides based on casual
observations of the rice crop and subjective assessments of damage or expected damage.
They rely mainly on chemicals to control rice pests because a prophylactic schedule
of applying insecticides with fungicides has been recommended for more than 20
years. Prophylactic measures are perceived as insurance against the threat of loss
from pest damage.
Pesticides applied in neighboring fields often prompt farmers to also apply pesticides
because they fear that the pests will migrate to their fields.
Limited knowledge of rice pests, lack of familiarity with scheduled applications,
anxiety to save the crop, lack of group efforts, and inadequate knowledge of IPM
methods have shaped farmers’ pest management. To improve farmers’ practices, the
focus must be on educating them on a pragmatic approach to reducing pest damage
by resorting to appropriate chemical and nonchemical methods of pest management
and promoting group efforts.
To increase the adoption of IPM practices, an efficient field-level organization of
farmers is indispensable for synchronizing their pest management practices through
groups (Goodell 1990). Group efforts also build the confidence levels of more farmers
more efficiently.
Educating rural children on growing crops and on pest management will encourage
the use of nonchemical control methods in the long run. Conducting IPM demonstrations
in farmers’ fields and working out the cost-benefit ratio will help to dispel
any negative attitudes farmers may harbor about these practices. Demonstrations also
enrich their knowledge and skills.
Pest management practices of rice farmers in Tamil Nadu, lndia 83

To feed the increasing population, government planners have given top priority
to finding ways to increase rice production while the land planted to rice is decreasing.
To achieve this, improved practices are needed to increase the yield per unit area.
But intensive cultivation and changing agroclimatic factors can often aggravate pest
problems, making farmers more anxious about protecting their crops. In this context,
farmers must adopt pragmatic pest management practices to cope with changing environments
and to ensure sustainable production. Achieving this objective is a challenge
because the farming community is vast and the majority of its members have
small, marginal holdings. Planners’ long-term policies must be in tune with these
facts to meet production goals.
References
Athimuthu P. 1990. Diagnostic study on information management, learning experience and
extent of adoption of nutrient use technology for rice. Ph.D. thesis. Tamil Nadu Agricultural
University, Coimbatore.
Gogoi SK, Gogoi DK. 1989. Adoption of recommended plant protection practices in rice: a
multivariate analysis. Indian J. Exten. Educ. 25(1&2):27-29.
Goodell G. 1984. Challenges to international pest management research and extension in the
Third World: do we really want IPM to work? Bull. Entomol. Soc. Am. 30(3):20-26.
Goodell G, Andrews KL, Lopez JI. 1990. The contributions of agronomo-anthropologists to
on-farm research and extension in integrated pest management. Agric. Syst. 32(4):321-
339.
Grieshop JI, Zalom FG, Miyao G. 1988. Adoption and diffusion of integrated pest management
innovations in agriculture. Bull. Entomol. Soc. Am. 34(2):73.
Keith DG, Linder DK, Wetzstern ME, Musser WN. 1983. Evaluating extension integrated pest
management programs with factor analysis indices. Entomol. Soc. Am. 29(3):43-46.
Kenmore PE. 1987. Criteria for the economic evaluation of alternative pest management strategies
in developing countries (cited in Waibel H and Engelhardt T). FAO Plant Prot. Bull.
36(1):27-33.
Mohankumar VA. 1985. An analysis of diffusion of selected paddy varieties in Chengelpattu
district. M.Sc. thesis. Tamil Nadu Agricultural University, Coimbatore.
Rajagopalan V. 1983. Deceleration of rates of agricultural growth in Tamil Nadu: trends and
explanatory factors. Indian J. Agric. Econ. 38(4):568-584.
Ramasamy C, Shanmugam TR, Suresh Kumar D. 1993. Constraints to higher rice yield in
different rice production environments and prioritization of rice research in Southern India.
Department of Agricultural Economics, Tamil Nadu Agricultural University,
Coimbatore. p 95-98.
Reichelderfer KH. 1980. Economics of integrated pest management: discussion. Am. J. Agric.
Econ. 62(5):1012-1013.
Sagar RL, Pal MK. 1986. Adoption of plant protection in major field crops as related to some
selected characteristics of farmers. Pest. Infor. 11(4):1-4.
Salvi PV, Gajre VG. 1985. Source of information used by the farmers in stages of adoption of
recommended farm practices. Pesticides 19(6):68-69.
Satapathy C, Pattnaik T. 1986. Awareness about plant protection measures. Pest. Infor. 11(4):6.
Seeta PK. 1985. The treatment of pesticides in the production function framework: a skeptical
note. Indian J. Agric. Econ. 40(2): 123-129.
84 Sivakumar et al

Sen D, Doijad PS. 1983. Factors influencing the adoption of integrated pest management and
agricultural productivity. Pesticides 17:3-7.
Shandta G. 1992. Integrated pest management in rice: achievements and opportunities. Ph.D.
thesis. Tamil Nadu Agricultural University. Coimbatore.
Subrahmanyam CK. 1985. Acceptance and adoption of TNAU innovations of paddy varieties.
M.Sc. thesis. Tamil Nadu Agricultural University, Coimbatore.
Teng PS, Savary S. 1992. Implementing the systems approaches in pest management. Agric.
Syst. 40:237-264.
Veerasamy S, Satapathy C, Appa Rao G. 1992. Constraints of rice production in coastal Orissa.
Indian J. Exten. Educ. 28(1&2):71-76.
Notes
Authors’address: Tamil Nadu Agricultural University. Coimbatore 641 003, Tamil Nadu, India.
Citation: Heong KL, Escalada MM, editors. 1997. Pest management of rice farmers in Asia.
Manila (Philippines): International Rice Research Institute.
Pest management practices of rice farmers in Tamil Nadu, lndia 85

CHAPTER 6
Pest management practices
of rice farmers in West Java,
Indonesia
S. Kartaatmadja, J. Soejitno, and I.P. Wardana
Surveys were carried out in the districts of lndramayu, Subang, and
Karawang on the north coast of West Java in 1991 and 1993. The
first survey was designed to evaluate farmers’ knowledge, attitudes,
and practices in controlling rice pests. A more in-depth survey of rice
pest management was conducted in 1993 in Desa village, Cikalong,
representing Karawang district. Farmers ranked the white stem borer
as the most important rice pest on the north coast of West Java,
followed by the brown planthopper. Most farmers agreed that rats
were the most severe rice pest during the dry season. Farmers reported
bacterial leaf blight, called kresek, as the most common rice
disease in the field. Bacterial red stripe or bacterial orange leaf blight,
a newly identified rice disease, was reportedly found in the field. In
both surveys, farmers’ reliance on insecticides as a control method
for rice pests was high. Surprisingly, most of them (76%) applied
broad-spectrum insecticides that are not recommended for use
against rice pests. Rice variety IR64 was widely used on the north
coast of West Java (82% in the wet season and 46% in the dry
season of 1991). But rice farmers in Cikalong avoided planting IR64
in both seasons to prevent white stem borer infestation. Farmers
collected egg masses of white stem borer in rice seedbeds.
Introduction
In 1976, rice farmers experienced a catastrophe when more than 500,000 ha of ricefields
were devastated by brown planthoppers ( Nilaparvata lugens ). Following the BPH
outbreak, the incidence of rice grassy stunt and ragged stunt virus hampered efforts to
increase rice production (Hibino et al 1977, Palmer and Rao 1981). In 1986, approximately
76,000 ha of ricefields in Central Java were again destroyed by BPH. which
led the government to ban broad-spectrum insecticides (Oka 1991a). In the wet season
of 1989-90, no fewer than 65,000 ha of ricefields were damaged by white stem
borers ( Scirpophaga innonata ) (Oka 1991 b). The following year, 7,000 ha of ricefields
87

in Karawang were again infested by white stem borer, resulting in zero rice yield.
Sudarmadji et al (1994) discovered that more than 10,000 ha of ricefields in Indramayu
were heavily attacked by white stem borer. This outbreak, apparently localized in
these districts, occurred during the early years of implementation of the integrated
pest management (IPM) program in Indonesia (Oka 1991a).
Banning 57 broad-spectrum insecticides along with the government policy to
eliminate insecticide subsidies marked the government’s effort to implement IPM
nationwide. But pesticides are believed to cause pest problems rather than to solve
them (Stern et al 1959). And we cannot simply tell farmers not to use insecticides to
protect rice plants from insect damage. Rice farmers in irrigated areas have been
participating in the rice production intensification program for a long time, in which
controlling rice pests with insecticides is mandatory. Escalada and Heong (1993) stated
that the use of pesticides has led farmers to associate pesticide use with modernism,
Efforts are being made to change these farmers’ attitudes and perceptions. The farmer
field school (FFS) (Oka 1991a) and farmer participatory research (FPR) (Escalada
and Heong 1993, Fujisaka et al 1993) are possible methods to be used to convince
farmers that insecticide use is not always necessary.
Studies on farmers’ behavior and state of knowledge have not been made in Indonesia.
Research on rice pest management is based mostly on what researchers think is
best for farmers. But this approach is no longer acceptable. Instead, farmer-driven
research has to be done before developing the research strategy. Therefore, the knowledge,
attitudes, and practices (KAP) of rice farmers in pest management should be
determined. Escalada and Heong (1992b) revealed that the slow adoption of IPM has
been attributed to the widespread gaps in farmers’ knowledge on rational pest management.
This paper presents a profile of farmers’ knowledge, attitudes, and practices in
rice areas where pest occurrence is high. In endemic areas, farmers tend to apply
excessive insecticides. Strategic research can be designed to examine the problem of
the insect pest complex arising from this situation. Where rice planting time is determined
by the allocation of irrigation water and resistant cultivars are not available,
farmers might rely on insecticides to control rice pests. Successful implementation of
IPM would depend on farmers’ understanding of it.
Methods
The data in these studies were collected using a questionnaire written in the Indonesian
language. The questionnaire was pretested before a formal survey was carried
out. The questionnaire was revised to avoid ambiguity, redundancy, and bias. One
farmer group in Ciasem Tengah village, Subang, was selected for the survey. In this
village, farmers have experienced white stem borer damage to their rice crop. The
first KAP survey was conducted in 1991 in three major rice-producing districts in
West Java. In 1993, a farmer participatory research, one approach to implementing
IPM, was conducted in Cikalong village. A KAP survey was conducted in this
88 Kartaatmadja et al

village before FPR began. The same questionnaire developed for the first survey was
used in the second one. Key data were analyzed using the chi-square test.
First survey
The districts of Indramayu, Subang, and Karawang on the north coast of West Java,
the main rice-producing areas, were selected as survey sites. The survey was conducted
in September-October 1991.
The total imgated rice area in these districts is 271,000 ha or 60.3% of the total
irrigated area in Jalur Pantura (north coast of West Java), which represents 30.8% of
that area in West Java (Fig. 1). Two agricultural extension agency working areas
(WKBPP) from each district were selected based on the high incidence of white stem
borer in the previous year. A WKBPP covers an area equivalent to a subdistrict. From
each WKBPP, two villages were selected (Table 1). Every village has 14 farmer groups,
each with 50–70 members. For the purpose of this survey, only two farmer groups
were selected. The respondents from each group were chosen randomly. Before the
survey, the field interviewers were briefed on the content of the questionnaire. They
were also asked to be neutral.
Fig. 1. Location of areas in West Java chosen as survey sites.
Pest management practlces of rlce farmers In West Java. Indonesia 89

Table 1. Villages selected as sites for the survey on knowledge,
attitudes, and practices, 1991.
District
lndramayu
Subang
Karawang
WKBPP a
Kebulen
Patral
Compreng
Ciasem
Pedes
Tegalsari
Village
Silyeg
Tarnbi
Anjatan
Cilandak Lor
Compreng
Mekarjaya
Sukahaji
Ciasem Tengah
Laban Jaya
Payung Sari
Kalibuaya
Cadas Kertajaya
a WKBPP = agricultural extension agency working area, which covers approximately
10,000 ha.
Second survey
Cikalong village in the district of Karawang was selected as the survey site because it
(1) has been chosen by local administrators as a promising village in the district,
(2) has high accessibility in terms of transportation and communication, (3) experienced
severe damage to the rice crop from white stem borer, and (4) has a high percentage
of farmers who have attended the farmer field school. All 446 farmers in the
village were members of seven farmer groups, and 47 were FFS alumni. A total of
128 farmers were selected randomly from all farmers who had not attended the FFS.
Likewise, all 47 FFS alumni and 6 FFS alumni from neighboring villages, but who
were members of farmer groups in Cikalong village, were included as respondents in
this survey.
Results
Demographic profile
The average age of farmers interviewed in Indramayu, Subang, and Karawang was
42.9. Most had 3–5 yr of schooling. But their experience in rice farming was relatively
long (17–22 yr). Some had experienced severe outbreaks of brown planthopper,
rice grassy stunt, and rice ragged stunt. The average landholdings in Indramayu,
Subang, and Karawang were 1.0, 1.1, and 1.5 ha, respectively. In the three districts,
the percentage of farmers who own and cultivate land was relatively higher than those
who share it with other farmers. The farmers’ profile was similar in these districts, as
it was in Cikalong village (Table 2).
Farmers perceptions of rice pests
In the first survey, farmers were asked about any rice pests and diseases they were
familiar with. Perhaps because the ricefields in Karawang, Subang, and Indramayu
had been severely damaged by white stem borer (WSB), farmers could easily recog-
90 Kartaatmadja et al

Table 2. Demographic profile of respondents in the survey on knowledge, attitudes, and practices,
1991 and 1993.
1991
1993
Specification lndramayu Subang Karawang Overall FFS a Non-FFS Overall
Age (yr) 53.5 40.9 44.3 46.2 42.1 44.2 43.1
Education (yr) 3.75.0 4.74.5 5.8 5.7 5.7
Experience in farming (yr) 18.9 17.0 21.5 19.1 19.5 22.2 20.8
Landholdings (ha) 1.0 1.1 1.5 1.2 0.9 1.3 1.1
Land tenure (%)
Owner 58.0 62.0 49.0 56.3 77.0 84.0 80.5
• Share 42.0 38.0 51.0 43.7 23.0 16.0 19.5
a FFS = farmers’ field school (SLPHT in Indonesian).
nize this infestation through the plant appearance, which showed deadheart and whitehead.
Under strong light, the WSB moth could be detected in houses or streets. We
were not able to measure farmers’ knowledge in differentiating BPH from whitebacked
planthopper (WBPH), simply because in the first survey the interviewers were not
provided with pictures of both insects. But farmers were quite familiar with hopperburn
as a symptom of rice plants damaged by BPH. Farmers referred to bacterial leaf blight
as kresek, bacterial red stripe as red disease, and gall midge as onion leaves of rice.
Using this approach, farmers were asked to name the rice pests and diseases they
considered important. They were allowed to list more than one rice pest and disease.
During the wet season, WSB was ranked as the most important pest (35%), followed
by BPH (31 %), bacterial leaf blight (31%), and rats (30%) in the three districts.
In the dry season, rats were considered as the major pest (65%), followed by bacterial
red stripe (42%) and WSB (18%). Farmers in Cikalong village were asked about
friendly insects or those that could help control rice pests. Having been shown pictures
of common predators of rice pests, farmers listed them in order of importance as
spiders, frogs, snakes, dragonflies, Ophionea sp. (Scarabidae), and Tenodera sp.
(Mantidae).
Rice varieties
First introduced in Indonesia in 1986, rice variety IR64 has become popular with
farmers. Farmers were asked to recall the varieties they planted during the past four
seasons, from the wet season of 1989-90 up to the dry season of 1991. Throughout
these seasons, IR64 appeared to be the dominant variety in Karawang, Subang, and
Indramayu. In the wet season of 1989-90, 89% of the respondents planted their fields
with IR64, and only 11% planted a variety other than IR64, such as Ciliwung, Cisadane,
IR42, and Way Seputih. The percentage of farmers planting IR64 in the dry season
was relatively lower than that in the wet season. About 42% of the farmers interviewed
in Karawang, Subang, and Indramayu had been growing IR64 for at least four
consecutive seasons. To our surprise, few farmers in Cikalong village grew IR64 during
the wet season of 1992-93 and dry season of 1993 (Fig. 2). A more diverse set of
rice varieties was found in Cikalong village, such as Muncul, Cisadane, and IR42.
Pest management practices of rice farmers in West Java, lndonesia 91

Fig. 2. Varieties cultivated by farmers in Cikalong, Karawang.
When farmers were asked why they chose a particular variety, both surveys revealed
that the high price of rice, resistance to pests and diseases, and high yields were the
most important considerations in deciding which variety to grow.
Physical and mechanical control
Following the outbreak of WSB in the wet season of 1989-90, it was recommended
that farmers cut and burn the standing rice plants to avoid further dispersal of the
white stem borer. School children collected egg masses, particularly those in seedbeds.
Among the farmers interviewed, 34% collected egg masses and 33% burned
rice straw. Only 8% of the farmers practiced trapping WSB with a kerosene lamp. The
situation was similar in Cikalong village. About 15% of the farmers trapped WSB
moths using kerosene lamps and another 15% burned rice straw. Some 49% of the
FFS alumni interviewed reported collecting WSB egg masses in seedbeds, whereas
only 17% of non-FFS alumni collected egg masses.
Chemical control
Nonrecommended insecticides. The 1991 survey revealed a high percentage of farmers
interviewed who applied broad-spectrum insecticides, which are not recommended
for use against rice pests. More than three-fourths of the farmers (76%) used recommended
and nonrecommended insecticides. Only 23% of the farmers used recommended
insecticides, and only 1% did not apply any insecticides. The survey in
Cikalong village showed that 63% of non-FFS alumni applied nonrecommended insecticides
and 23% of FFS alumni were still using the same group of insecticides. A
survey conducted by Central Policies for Implementation Studies (CPIS) (Suyanto et
92 Kartaatmadja et al

al 1994) in some districts of Java revealed that FFS farmers (25%) were no different
from non-FFS farmers (29%) in their use of nonrecommended insecticides.
Time of application. Some 62.5% of the farmers in Karawang, Subang, and
Indramayu applied insecticides at least once in seedbeds. The preferred time for farmers
to apply insecticides was during the first 30 d after transplanting. Some 36-86% of
the farmers followed this calendar system. The 1993 survey showed a similar pattern,
in which FFS farmers (75%) and non-FFS farmers (82%) applied insecticides in seedbeds.
The application of insecticides by both groups of farmers during this stage was
not significant. During the first 30 d after transplanting, 86% of non-FFS farmers
applied insecticides to their ricefields, a much higher percentage than their FFS counterparts
(Fig. 3). The survey by CPIS (Suyanto et a1 1994) showed that 57% of FFS
farmers and 60% of non-FFS farmers sprayed insecticides during the first 30 d after
transplanting.
Frequency of application. The 1991 survey showed that farmers applied carbofuran
in seedbeds and in fields. Other insecticides used were BPMC and organophosphate.
A clearer picture of the frequency of farmers’ insecticide applications emerged in
Cikalong village (Fig. 4). The statistical analysis showed that the non-FFS alumni
applied insecticides more than four times in one cropping season, which is significantly
higher than what the FFS alumni did. Few non-FFS alumni applied fewer than
two times. in contrast to FFS alumni, who preferred to apply once or not at all in a
cropping season. The same percentages of FFS and non-FFS alumni, however. frequently
apply insecticides two to three times per season.
Fig. 3. Time of insecticide application for rice in Cikalong,
wet season 1992-93.
Pest management practices of rice farmers in West Java. Indonesia 93

Fig. 4. Frequency of insecticide application in Cikalong, wet
season 1992-93.
Discussion
The districts of Karawang, Subang, and Indramayu are among 20 districts in West
Java. These three districts may not represent the agricultural situation in the province,
especially with respect to knowledge, attitudes, and practices of rice farmers in pest
management. But almost certainly the three districts represent the seven districts on
the north coast of West Java. The main difference between districts on the north coast
and in the central plain or southern part of West Java is the occurrence of rice pests
and diseases. More frequent outbreaks have been reported on the north coast compared
with almost none in the southern part of West Java (Sudarmadji et al 1994).
Results for these three districts, however, may reflect similar situations in other provinces
having the same problems with rice pests and diseases.
The two surveys focused on issues that might affect farmers’ decision making in
rice pest management: (1) perceptions of pests and diseases, (2) knowledge of the
importance of predators, (3) goals in the farming system, (4) perceptions of control
practices recommended by authorities, and (5) views on the use of chemical control,
particularly insecticides.
In the past, much attention was given to BPH control and even policies to implement
IPM were mostly based on the BPH experience. Much effort was directed to
developing rice varieties resistant to BPH and banning 57 broad-spectrum insecticides.
This is shown by the outbreak of white stem borer on the north coast of West
Java in early 1990, which devastated 65,000 ha of rice.
94 Kartaatmadja et al

When the survey respondents were interviewed in 1991, most farmers mentioned
WSB as an important pest during the wet season. The response might have been
different if farmers had been asked before the WSB outbreak. Farmers’ knowledge of
rice pests and diseases and how they perceive these in relation to yield loss is unclear.
It has been shown that they equate the importance of bacterial leaf blight to that of
BPH and WSB, which in fact was not always the case. Bacterial leaf blight may, to
some extent, cause reduced yield, but this may not be as high as the yield loss caused
by BPH and WSB. It seems that farmers use visible plant symptoms as a criterion for
ranking pest importance. Noticeable predators would attract farmers’ attention, such
as snakes as predators of rats.
Farmers on the north coast of West Java are usually commercial rice farmers. For
example, interviewed farmers (FFS and non-FFS alumni) mentioned the high price of
rice as their most important criterion in deciding what variety they should grow. Because
of its resistance to BPH, along with other preferred varietal characteristics,
IR64 has been widely accepted by farmers in West Java. But once farmers found that
IR64 was more susceptible to WSB, they switched to other varieties that offered more
reliable yields. Only a few farmers in Cikalong planted IR64. Many Indonesian farmers
clearly considered resistant rice varieties as an essential component of IPM. Therefore,
breeding for resistance to the main pests and diseases will remain important.
Farmers’ reluctance to destroy rice plants infested with WSB or BPH is understandable.
Those farmers who have suffered severe yield losses could not afford to
spend more money to cut and burn rice straw, and they did not see a direct advantage
for themselves. Collecting egg masses to control WSB might be useful when done in
seedbeds and when the number of egg masses is quite low. Rauf et al (1992) revealed
from their study in Karawang, West Java, that kerosene lamps were not effective as a
tool to trap the WSB moth because these lamps trapped only a few moths. Some of
the trapped moths had already laid their eggs in the field before entering the lamp.
The depth of farmers’ knowledge and practices of pest management can be measured
in a number of ways. The variables could be an appreciation of natural enemies,
the ability to detect and identify a pest problem, or attitudes toward insecticide applications.
The National IPM Program in Indonesia promotes the conservation of natural
enemies. To enhance this natural biological control, pesticide application has to be
minimized and, if possible, stopped. It is possible that in SLPHT (FFS), farmers did
not receive adequate knowledge or gain enough experience in using and handling
insecticides. Suyanto et al (1994) used insecticide knowledge and practices as indicators
of the level of farmers’ knowledge of IPM. Their finding is somewhat astonishing:
among FFS farmers interviewed, 55–60% of them used insecticide sprays at least
once during the first 30 d after transplanting, and 51–58% of those who sprayed used
a nonrecommended insecticide, that is, approximately 26% of the FFS farmers interviewed.
This result was consistent with findings obtained from surveys conducted in
Karawang, Subang, and Indramayu. Similar findings were reported by groups of researchers
from IPB (1995), RILET (1996), and RIR (1996). The question is why
these findings differ from those reported by the National IPM Program (1993). The
discrepancy may be due to differences in the survey methods.
Pest management practices of rice farmers in West Java, lndonesia 95

Both surveys indicated that most farmers in an endemic area did not apply insecticides
prophylactically. When queried why he applied insecticides without waiting
until the pest population reached the economic threshold, one farmer said that he was
womed that all rice pests from the neighboring areas would move into his field (Soejitno
1991). Farmers’ negligence in the use of recommended insecticides has been reported
to occur in other places (Kartaatmadja et al 1991). Farmers might not be aware of the
weaknesses of broad-spectrum insecticides as a possible cause of insect resurgence
and hazards to beneficial insects. Farmers’ continuous use of broad-spectrum insecticides
could trigger the buildup of other insect pests, particularly BPH.
The high percentage of rice farmers who applied broad-spectrum insecticides,
including those who attended the FFS, can be explained by the following: (1) WSB
was considered as their most important rice pest, (2) the first survey was conducted
just a year after the outbreak, and (3) this group of insecticides is available in the
market.
To improve pest management, farmers must have the knowledge and skills to
diagnose pests and diseases, assess the presence of natural enemies, and decide on
appropriate control measures. The FFS was designed to meet this objective. Apparently
farmer adoption of the IPM concept is not only determined by the degree of
farmers’ knowledge but also by how they view IPM. Researchers at IPB (1995) demonstrated
that the low adoption of IPM by FFS alumni is due to the rejection of the
IPM concept itself by farmers, because they were not confident about implementing
it. Norton and Heong (1988) stated that farmers might not adopt IPM because it will
not make them better off.
Winarto (1996) pointed out that a simple analysis of pest management by trainers
could lead to a simple conclusion, that is, to spray or not spray, instead of a general
crop management strategy. Farmers could not appreciate that the absence of predators
was caused by the spraying of insecticides. In her study in Ciasem subdistrict, Subang,
Winarto (1996) mentioned three possible reasons why FFS alumni did not practice
what they learned from their trainers.
1. Training itself is too scientific. The training began in grade three instead of
grade one.
2. Farmers believed that FFS and IPM knowledge originated from the government
and scientists who emphasized pests and diseases and how to control
them, rather than the broader problems of rice cultivation. Farmers perceived
the FFS as only a school about pests and diseases.
3. Trainers did not carry out any cultivation practices along with the farmers to
demonstrate the validity of the new approach.
96 Kartaatmadja et al

References
Escalada MM, Heong KL. 1993. Communication and implementation of change in crop protection.
In: Crop protection and sustainable agriculture. Chichester: John Wiley and Sons.
p 191-207.
Fujisaka S, Guino R, Medrano P, Obusan L. 1992. Establishing a farmer participatory pest
management experiment. Paper presented at the ARFSN-INSURF-IPM Joint Meeting,
12- 17 Oct 1992, Ho Chi Minh City, Vietnam. 11 p.
Hibino H, Roechan M, Sudarisman S, Tantera DM. 1977. A virus disease of rice (kerdil hampa)
transmitted by the brown planthopper in Indonesia. Contributions 35:1-15.
IPB (Bogor Agricultural University). 1995. Characteristic relationship of Farm Field School
with knowledge and adoption gap of IPM in selected villages in Indramayu district, West
Java. National IPM Project. 113 p. (In Indonesian.)
Kartaatmadja S, Baehaki SE, Suparyono, Swastika dan DKS. 1991. The status of rice pests and
diseases at farmer’s level. A case study in WKBPP Sumbang and Jompo Kulon. Banyumas.
Reflector 4:34-36.
National Integrated Pest Management Program. 1993. The impact of IPM training on farmer
behavior: a summary of results from the second field school cycle. National IPM Program.
42 p.
Norton GA, Heong KL. 1988. An approach to improving pest management of rice in Malaysia.
Crop Protect. 7:84-90.
Oka IN. 1991 a. Success and challenges of the Indonesian National Pest Management Program
in a rice-based cropping system. Crop Protect. 20: 163.165.
Oka IN. 1991b. Assessment of control for rice stem borer based on IPM concepts. National
IPM Program. 19 p.
Palmer LT, Rao PS. 1981. Grassy stunt, ragged stunt, and tungro diseases of rice in Indonesia.
Trop. Pest Manage. 27:212-217.
Rauf A, Santosa S, Nurmansyah A, Santosa TH. Rudianto dan SA. 1992. Spatial and temporal
development of white stem borer, Scirpophaga innotata and its implication for research
strategy. National IPM Program. 9 p. (In Indonesian.)
RILET (Research Institute for Legumes and Tuber Crops). 1996. Evaluation of farmer field
school impact, institutional involvement and farmer participation in IPM implementation
in East Java. National IPM Project. 115 p. (In Indonesian.)
RIR (Research Institute for Rice). 1996. Social and economic impact of IPM implementation
in West Java. National IPM Project. 100 p. (In Indonesian.)
Soejitno J. 1991. Biology and control of rice stem borer. In: Soenarjo E, editor. Rice. Central
Research Institute for Food Crops. Bogor (Indonesia): Book 3:713-735. (In Indonesian,)
Stern VM, Smith RF, Van den Bosch R, Hagen KS. 1959. The integrated control concept.
Hilgardia 29:81-101.
Sudarmadji, Rahayu A. Kusdiaman D, Suharto H, Wardana dan P. 1994. Rice plants damaged
by white stem borer Scirpophaga innotata in Indramayu in the wet season 1993/1994.
Reflector 7:21-25.
Suyanto, Hariyadi B, Budiyati S, Quizon J. 1994. Insecticide use in rice farming on Java: a
preliminary study comparing the behavior of SLPHT and non-SLPHT farmers. Agriculture
Group Working Paper No. 20. Centre for Policy and Implementation Studies. 25 p.
Winarto YT. 1996. Farmers’ perspectives on integrated pest management. Agric. Res. Ext. Newsl.
34: 16-20.
Pest management practices of rice farmers in West Java, Indonesia 97

Notes
Authors’ addresses: S. Kartaatmadja, Central Research Institute for Food Crops, Bogor 16111,
Indonesia. J. Soejitno and I.P. Wardana, Research Institute for Rice, Sukamandi, Subang
41256, Indonesia.
Acknowledgments: The authors wish to gratefully acknowledge the Swiss Agency for Development
and Cooperation (SDC) for supporting the first survey through the Rice IPM Network
coordinated by IRRI, Philippines. The second survey was funded by the Government
of Indonesia.
Citation: Heong KL, Escalada MM, editors. 1997. Pest management of rice farmers in Asia.
Manila (Philippines): International Rice Research Institute.
98 Kartaatmadja et al

CHAPTER 7
Pest management practices
of rice farmers in the rainfed lowland
environment of the Lao PDR
H.R. Rapusas, J.M. Schiller, and V. Sengsoulivong
A survey of rice farmers was conducted in 11 provinces in the northern,
central, and southern regions of Laos. A total of 463 respondents
selected randomly from the sites were interviewed in February,
March, and October 1994.
Most of the respondents were 30-45 years old and were owneroperators
(90.3%). Farm sizes ranged from 0.25 to 7 ha. Thirt-eight
percent had 1-2 ha, and a few had >5 ha of
land cultivated with rice. Farmers grew the traditional photoperiodsensitive
and glutinous varieties, growing 3–4 varieties in one season.
The cropping pattern was usually rice-fallow (75.6%), although
23% had rice-rice and 1% had rice-others (maize, vegetables, watermelon).
Stem borers, grasshoppers, rice bugs, gall midge, and other
lepidopterous larvae and worms were reported to be the most common
pests observed in fields. About 75% of the respondents indicated
that they did nothing to control these pests, whereas 22.7%
said they applied pesticides. Most of those who used pesticides
applied them only once, mostly in the seedbed and within the first
40 d after transplanting. Most of the pesticides used belong to categories
I and II. Yields of farmers who did not apply pesticides were
higher than for those who treated their crops against insects, rats,
and crabs.
Seventy-two percent of the respondents perceived that leaf-feeding
insects cause a yield loss and 43.5% believed that these insects
should be controlled early. Some respondents were aware of the
presence of natural enemies in their fields, such as spiders, dragonflies,
beetles, and wasps, but were uncertain about their roles.
99

Introduction
Laos is located almost entirely in the Mekong River watershed, with a total land area
of about 237,000 km 2 . Lowland alluvial plains and terraces in the central and southem
areas constitute 20%, hills backing the lowlands, 50%, and mountains of 1,000-
3,000 m above sea level, 30%, to the north and along the eastern border with Vietnam.
The country has a population of about 4.6 million.
As the most important sector of the economy of the Lao PDR, agriculture accounted
for about 56% of the total value added in 1993 (UNDP 1994) and involved
about 80% of the population. Rice is the most important crop, with about 83% of the
649,000 ha under cultivation devoted to rice. The main nonrice crops are maize (27,000
ha), vegetables (15,000 ha), and tuber crops (14,000 ha); the rest are cotton, soybean,
sesame, peanut, tobacco, and sugarcane, which collectively account for about 100,000
ha. Coffee and tea are grown on about 22,000 ha. Approximately 97% of the total
cultivated area depends on rainfall.
Total annual rice production ranges from 1.5 to 1.6 million t. But the dependence
on rainfall can result in annual fluctuations. In 1993, serious flooding reduced production
to 1.46 million t, whereas in 1995 it fell to about 1.25 million t as a result of
serious drought in much of the rainfed lowland environment (LAO-IRRI Project Report
1996, Lao PDR 1996). Most of the rice produced is consumed domestically, with
less than 5% believed to be traded.
The wet-season rice crop is produced on about 98% of the rice area and accounts
for 97% of production. Sixty-two percent of the rainfed lowland contributes 76% of
production, whereas 36% of the rainfed uplands accounts for 22% of production. In
1994, irrigated rice was grown on an estimated 11,000 ha and in 1995 on 13,300 ha.
The highest production from lowland rice comes from the central region (247,829 t)
and southern region (123,325 t). The northern region produces upland rice.
Approximately 86% of the rainfed lowland rice area is in the central and southern
agricultural regions, mainly in the provinces adjacent to the Mekong River. The
major rice-producing plains are the Vientiane Plain (Vientiane Province and Vientiane
Municipality), Borikhamxay, Sebang-Faay (Khammouane and Savannakhet), Sebang-
Hiang (Savannakhet Province), Sedone (Saravane Province), and Champassak. The
topography of this area mainly consists of a system of ancient low terraces with an
elevation of about 200 m. The remaining rainfed lowlands mostly occupy rather narrow
river valleys in the more northern provinces. Some production areas of between
500 and 2,000 ha are also found in the north.
Rainfall in most provinces along the Mekong River valley ranges from about
1,500 to 2,200 mm on average per year, with about 75% being received from May to
October. In other northern provinces (Sayabouly and Luang Prabang), it drops to
about 1,300 mm. August and September are considered the wettest months of the
year, when heavy rains can result in localized flooding. Soils of much of the riceproducing
area adjacent to the Mekong River are derived mainly from old alluvial
deposits and sandstone materials in the provinces of Saravane and Savannakhet.
100 Rapusas et al

In the rainfed lowland environment, a single wet-season rice crop is the predominant
agricultural activity. Farmers commonly plant traditional glutinous photoperiodsensitive
varieties, with each farmer growing 34 varieties of varying maturity periods.
Early maturing varieties are normally planted in the upper terraces, where water
supply is less certain, whereas later maturing varieties are grown in the lower areas.
Early maturing cultivars constitute 20% of the varieties planted, whereas medium
varieties make up 50% and late varieties 30%. Until about 1993, the rate of adoption
of improved varieties was low and based mainly on the use of the Thai glutinous
varieties RD6, RD8, and RDl0 and nonglutinous variety Khaw Dok Mali 105 or
KDML 105. In 1993, however, the recommended Lao varieties Niaw Thadokham 1
(TDK1), Niaw Thadokham 2 (TDK2), and Phone Ngam 1 were released and subsequently
adopted by farmers, particularly in central and southern Laos.
The usual cropping calendar is for seedbed rice to be sown in early to mid-June
and transplanted in early to mid-July. Timing of these operations, however, is much
dependent on rainfall distribution. Because the rice production cycle is at a very low
level of mechanization, more than 95% of the area is cultivated using traditional buffalo-drawn
plows. Planting, weeding, and harvesting are all done manually. Mobile
threshers have become evident since 1993, but they still account for less than 10% of
the harvested area; the remainder is threshed manually.
The level of technology adoption in the country is usually low. Apart from family
labor, farm inputs in the system are minimal. Organic fertilizer in the form of farm
yard manure is usually applied to seedbed rice.
The government’s aim of raising total production to between 2.0 and 2.2 million
t by the year 2000 is largely based on achieving a significant production improvement
in the rainfed lowland environment through improved cultural practices, nutrient inputs,
and integrated pest management.
Pesticide use
Lao farmers do not have a history of pesticide use. This reflects a combination of
active government discouragement of their use, a lack of access to them in much of
the country, and the limited purchasing power of small farmers. The limited use of
pesticides has been largely confined to areas of irrigated rice in provinces adjacent to
the Mekong River. But more recently, pesticide use has increased in the areas of vegetable
cultivation near the main population centers. Most pesticides used are insecticides
and rodenticides. In the late 1980s and early 1990s, some pesticides were brought
into the country in the form of development aid. In 1993, the Department of Agriculture
and Extension approved the import of approximately 63 t of various pesticides.
The relatively open borders with Thailand, Vietnam, and China encourage the entry
of pesticides, even those without government approval. Procedures for pesticide registration
and regulation have not been established yet.
Farmers’ perception of insect pests as a production constraint
A survey of farmer perceptions of the main production constraints in the rainfed lowland
was conducted in 1993 (Khotsimuang et al 1995). Farmer respondents (191)
Pest management practices of rice farmers in Lao PDR 101

Table 1. Farmer ranking of most important production constraints in
the rainfed lowland environment.
Province
District
Ranking of productton constraints
1 2 3
Vientiane Mun.
Vientiane
Khammouane
Savannakhet
Saravane
Champassak
Champassak
Sayabouly
Nasaythong
Thurakhom
Thakhel
Champone
Vapi
Sanasomboune
Phonethong
Phiang
Drought
Insects
Drought
Drought
Drought
Drought
Drought
Drought
Weeds
Drought
Insects
Crabs/snails
Insects
Insects
Insects
Insects
Insects
Weeds
Weeds
Weeds
Weeds
Weeds
Weeds
Weeds
Source: Khotsimuang et al 1995.
from nine districts in seven provinces were asked to rank 11 potential production
constraints, which included insect pests, rodents, crabs and snails, drought, weeds,
diseases, soil fertility, labor, varieties, credit, and flooding. Insect pests were rated
within the top three constraints for seven of the eight districts (Table 1), whereas
damage caused by crabs or snails was rated as second in importance in two districts.
Survey of farmer pest management practices
Farmers’ perceptions of pests, techniques available, and resources often influence the
adoption of improved pest management practices. An assessment and understanding
of farmers’ current practices, perceptions, and constraints is important for improving
their pest management decision making (Lim and Heong 1984, Mumford and Norton
1984, Norton and Heong 1988). Similarly, a better understanding of the biological
and ecological processes and socioeconomic factors that influence their decision
making is often necessary. Understanding farmers’ perceptions may have research
implications for improving pest management practices and decision-making processes.
Between February and March and in October 1994, a survey of farmers’ knowledge,
attitudes, perceptions, and current pest management practices was undertaken
in the rainfed lowland rice environment in 18 districts distributed over 11 provinces
of Lao PDR. Although some of the areas surveyed are irrigated during the dry season,
they are usually considered rainfed during the wet season because of the inability to
control water during this part of the year. The provinces included in the survey represented
all three regions of the country (Fig. 1). In the north were the provinces of
Luang Namtha, Oudumxay, Luang Prabang, and Sayabouly; in the central region
were Vientiane, Xieng Khouang, Borikhamxay, Savannakhet, and Vientiane Municipality.
The southern region was represented by the provinces of Saravane and
Champassak.
102 Rapusas et al

Fig. 1. Survey sites ( • ).
Pest management practices of rice farmers in Lao PDR 103

Method
A total of 463 farmers were interviewed; individual respondents were selected at
random within each district (Table 2). The survey sites were identified by Lao researchers
and technicians who were collaborators of the LAO-IRRI Project and these
were mostly the places where they worked with farmers. From each district, two or
three villages were selected for the survey. Respondents were selected by taking one
farmer for every five households unless the farmer was not available during the time
of the survey, in which case we chose the next farmer. We conducted the interviews
using a structured questionnaire translated into the Lao language, and obtained information
related to cropping activities during the 1993 wet season. We coded, summarized,
and tabulated the gathered data in frequency tables.
Sociodemographic profile
Farm size and tenure status. The total planted area of respondents was about 756 ha,
producing 1,482 t of rice. Farm sizes ranged from 0.25 to 7 ha. The smallest landholdings
were in Oudumxay and Xieng Khouang (av 0.89 and 0.86 ha, respectively); the
largest were in Savannakhet (av 2.32 ha) and Vientiane Municipality (av 1.87 ha).
Some 38% of the respondents had less than 1 ha, and approximately 43% had between
l.1 and 2 ha. Only four respondents had more than 5 ha.
Table 2. Distribution of respondents.
Province District Respondents
(n=463)
Northern region
Luang Namtha
Oudumxay
Luang Prabang
Sayabouly
Xieng Khouang
Central region
Vientiane Mun.
Vientiane Prov.
Borikhamxay
Savannakhet
Southern region
Saravane
Charnpassak
Total
Xieng
Namtha
Xay
Luang Prabang
Phiang
Bonneau
Nasaythong
Saythany
Phonehong,
Thurakum
Paksan
Xaybouly
Champone
Saravane
Vapi
Champassak
Pakse
Sanasomboun
12
13
25
35
25
25
58
65
32
10
49
21
14
11
25
13
30
463
104 Rapusas et al

Some 90.3% of the respondents were owner-operators, mostly from the provinces
of Luang Namtha, Oudumxay, Xieng Khouang, Borikhamxay, Saravane, and
Xaybouli District of Savannakhet Province.
Age classes. Respondents’ ages ranged from 20 to 75, with farming experience
from 1 to 60 years. Most (61%) were between ages 30 and 45.
Production practices
Cropping pattern. Some 75% of the respondents planted a single wet-season rice
crop. Some farmers (23%) followed a rice-rice cropping pattern. Fewer than 1% reported
that they planted nonrice crops such as maize, watermelon, cucumber, and
vegetables after their wet-season rice.
Rice varieties. Farmers grew a wide range of varieties, most of which were traditional
glutinous photoperiod-sensitive varieties. Most households cultivated 3–4 different
varieties with varying maturity periods, to fit consumption needs and help distribute
the labor requirement. Seed supplies for most farmers (90%) came from their
own harvests or, occasionally, through seed exchange with other farm households in
the area. Those with access to research stations (10%) sometimes obtained some seed
from these stations. The most common methods of cleaning seeds before planting
were winnowing and flotation.
Fertilizer use. Of 135 respondents from the northern region, only 13.3% applied
inorganic fertilizer during the 1993 wet season. In contrast, 32% and 33% of the
respondents from the central and southern regions, respectively, reported the use of
fertilizer. Respondents using fertilizer were high in Vientiane Province (81%),
Savannakhet (80%), Saythany District of Vientiane Municipality (67%), Saravane
(66%), and Champassak (64%). None of the respondents from the provinces of Luang
Namtha and Xieng Khouang reported using fertilizer.
Pest management knowledge, practices, and perceptions
In the 1993 wet season, the five most commonly mentioned pests that attacked rice
plants were stem borers, orthopterans, rice bugs, gall midge, and leaffolders.
The importance of these pests varied among the sites (Table 3). In Luang Namtha,
grasshoppers and locusts, leaffolders, and caterpillars or worms were the three commonly
mentioned pests, whereas in Oudumxay they were grasshoppers, caterpillars,
and gall midge. Luang Prabang and Sayabouly respondents reported stem borers as
the most common pest, followed by gall midge, grasshoppers, and rice bugs. In
Vientiane Municipality, the two sites apparently had different pest problems. In
Nasaythong, rice bugs, stem borers, and grasshoppers were ranked as the three most
important pests, whereas in Saythany these were grasshoppers, stem borers, and thrips.
Crabs were ranked as fourth at both sites. In Vientiane Province, stem borers and rice
bugs were reported most. Stem borers and gall midge were reported in Borikhamxay
and Xaybouly of Savannakhet, whereas in Champone thrips and crabs were mentioned
most. In Champassak, gall midge, hoppers, and thrips were mentioned most.
Pest management practices of rice farmers in Lao PDR 105

Table 3. Pests that attack rice plants as reported by respondents.
Farmers reporting (%)
Vientiane Mun
Pest Luang Oudumxay Luang Sayabouly Xieng
Namtha Prabang Khouang Nay Say
Leaf feeders
Armyworm 0 0 0 3 0 7 0
Cutworms/worms 5 11 6 3 0 5 3
Caseworms 0 0 1 0 0 5 2
Leaffolder 9 6 4 7 0 5 1
Whorl maggot 0 0 0 0 0 0 0
Rice skipper 3 0 0 1 0 0 0
Grasshoppers/locust 20 23 19 12 0 23 16
Thrips 0 0 5 1 0 4 10
Brown planthopper 0 3 3 2 0 2 8
Whitebacked
planthopper 0 6 0 0 0 0 0
Green leafhopper +
zigzag leafhopper 0 2 2 1 0 1 0
Stem feeders
Stemborers 0 7 25 24 0 23 15
Gall midge 0 10 9 15 0 0 0
Grain feeders
Rice bugs 1 1 14 4 0 28 7
Mole crickets 0 3 3 1 0 0 4
Ants/termites 0 0 0 0 0 2 0
Other pests
Rats 1 0 3 4 0 14 9
Crabs 0 0 3 0 0 6 6
Birds 0 0 1 1 0 1 3
Diseases (blast,
sheath blight, 0 0 0 0 0 0 0
false smut)
Pest control practices
Pest control methods. Although respondents reported observing several pests in their
fields, 74.5% did nothing to control these pests. Some 22.7% used pesticides to control
them. A few others (2.8%) practiced handpicking, water management, baiting,
and indigenous methods such as using leaves of some trees (Gliceridia sepium and
Azadirachta indica, etc.) to control pests.
Pesticide use patterns. Of the 105 respondents who applied pesticides during the
1993 wet season, 75.2% of them made only one application and 20% made two applications.
A few farmers (3.8%) applied pesticides as many as three times. One farmer
from Savannakhet said he applied chemicals 10 times to control crabs.
Seventy-four percent of those who used pesticides applied them during the first
40 d after transplanting (DT). The other applications were either made in seedbeds or
after 40 DT.
106 Rapusas et al

Farmers reporting (%)
Savannakhet
Champassak
Vientiane Borikhamxay Saravane Total
Prov. Xay Cham Champ. Pakse Sana
0
0
0
0
3
0
4
1
3
0
0
0
0
0
0
1
0
0
8
5
0
13
0
1
8
3
3
2
3
1
3
0
0
0
9
0
0 0 0 0 20
7 0 3 4 55
0 5 0 0 16
17 3 0 0 68
0 1 0 0 4
0 0 0 0 5
3 0 0 0 129
0 3 6 10 52
19 0 2 14 59
0 0 0 0 0 2 1 7 16
0 0 0 0 0 3 0 0 10
25 8 30 2 6 3 3 2 173
4 8 22 3 1 5 0 20 97
24 2 8 1 6 0 2 2 100
2 0 0 0 0 0 0 0 13
0 1 0 0 0 0 0 0 4
0 0 0 0 0 1 0 0 33
0 5 20 8 1 0 2 8 59
0 0 0 0 1 0 0 0 7
0 0 0 2 1 0 0 1 4
Target pests. Pesticide applications in seedbeds were mainly for the control of
grasshoppers (Table 4). Some 54% of the applications during the first 40 DT were
aimed at controlling stem borers, crabs, caterpillars or worms, and hoppers. But 27%
of the farmers reported that they sprayed their plants as a prophylactic measure. Rice
bugs, stem borers, and rats were the targets of applications made after 40 DT.
Common pesticides used. Respondents indicated that methyl parathion was the
most commonly used insecticide. Others mentioned carbaryl, diazinon, carbofuran,
monocrotophos, metamidophos, endosulfan, BHC, and DDT. For rat control, farmers
used zinc phosphide. These pesticides belong to categories I and II. Table 5 shows the
pesticides used by respondents from the different provinces.
Precautionary measures in using pesticides. When farmers were asked what precautionary
measures they practiced when using pesticides, the most commonly mentioned
precaution was not to spray at all (27%). Some 12% were aware that wearing
protective clothing (e.g., long trousers, long sleeves) and covering the mouth and
Pest management practices of rice farmers in Lao PDR 107

Table 4. Target pests of pesticide applications a .
Farmers reporting (no.)
Seedbed 1st 40 DT 41-60 DT >60 DT
(n=11) (n=79) (n=11) (n=4)
Pests Total %
No. % No. % No. % No. %
Grasshoppers 7 64 8 10.0 1 9.1 0 0 16 14.9
Caterpillars/worms 1 9 13 16.5 1 9.1 0 0 15 14.0
BPH a 1 9 11 14.0 0 0 1 25 13 12.1
Leaffolders 00 4 5.0 2 18.2 1 25 7 6.5
Stemborers 0 0 26 33.0 3 27.3 0 0 29 27.1
Gall midge 00 4 5.0 0 0 0 0 4 3.7
Rice bugs 00 3 3.8 4 36.4 2 50 9 8.4
Thrips 00 7 8.9 0 0 0 0 7 6.5
Crickets 00 2 2.5 0 0 1 25 3 2.8
Others 00 6 7.6 0 0 1 25 7 6.5
Crabs 1 9 14 17.7 1 9.1 1 25 17 15.9
Rats 00 1 1.3 0 0 3 75 4 3.7
General protection 1 9 21 26.6 0 0 0 0 22 20.5
for any pest
a DT = days after transplanting, BPH = brown planthoppers.
nose (10%) while spraying were important. Ten percent were also aware that eating
or smoking should be avoided while spraying. Other precautionary measures mentioned
were not entering sprayed fields (5%), taking a bath or washing after spraying
(8%), and using IPM, reading labels carefully, asking for advice, keeping chemicals
away from children and animals, not throwing leftover solutions in water canals or
rivers, not eating fish or vegetables from sprayed fields, and several others (14%).
Source of plant protection advice. For plant protection advice, the agricultural
officers of the respective provinces were the major source of information (53.5%).
Farmers also obtained information from other farmers. either their neighbors or from
other places (15%). Pesticide dealers (1%), relatives (2%), radio (2.5%), television
(l%), newspapers (l%), and some personnel from NGOs and banks were also mentioned.
Some respondents (27.5%) claimed that they depended on their own knowledge
and experience.
Farmers’ perceptions
Perceptions on effectiveness of pesticides. When respondents were asked whether or
not pesticides could increase their yield, 68% believed that these pesticides would.
Few (8%) said that pesticides would not increase yield, whereas others (23.9%) had
no opinion.
Toxicity of pesticides. The majority of the respondents (72%) believed that pesticides
would kill humans, fish, and animals. Respondents who had not used insecticides,
such as those from Xieng Khouang, had no idea about the effect of pesticides
on humans, fish, and animals.
108 Rapusas et al

Farmers (no.) using pesticides by site
Pesticides used Luang Oudum- Luang Saya- Xieng Vientiane Vientiane Borikham- Savanna- Saravane Champassak Total %
Namtha xay Prabang bouly Khouang Mun. Prov xay khet
Zinc phosphide 0 0 0 0 0 2 0 0 0 0 0 2 1.8
Table 5. Pesticides used by respondents.
Organophosphate
Methyl parathion 0 0 8 0 0 11 4 0 9 0 8 40 37.4
(Folidol)
Diazinon 2 1 3 1 0 1 0 1 1 2 0 12 11.2
Monocrotophos 0 0 0 0 0 0 6 0 1 2 4 13 12.1
(Azodrin/
Nuvacron)
Metharnidophos 0 0 0 0 0 3 0 0 0 0 0 3 2.8
(Tamaron)
Endosulfan 0 0 0 0 0 3 0 0 0 0 0 3 2.8
(Thiodan)
BHC (Lindane) 0 0 0 1 0 0 0 0 0 0 0 1 0.9
Carbamate
Carbofuran 0 0 0 0 0 0 2 0 3 0 4 9 8.4
(Furadan)
Carbaryl (Sevin) 0 0 15 2 0 1 3 0 2 1 1 26 24.3
Rodenticide

Table 6. Farmers’ perceptions on whether all arthropods in their
ricefield damage rice plants.
Farmers reporting (no.)
Province Yes No Do not know
No. % No. % No. %
Northern region
Luang Namtha 6 1.3 14 3.0 5 11
Oudumxay 2 0.4 22 4.8 1 0.2
Luang Prabang 6 1.3 20 4.3 9 1.9
Sayabouly 22 4.8 3 0.6 0 0
Xieng Khouang 0 0 1 0.2 24 5.2
Central region
Vientiane Mun. 25 5.4 53 11.4 45 9.7
Vientiane Prov. 10 2.2 12 2.6 10 2.2
Borikhamxay 00 7 1.5 3 0.6
Savannakhet 8 1.7 34 7.3 28 6.0
Southern region
Saravane 7 1.5 17 3.7 1 0.2
Champassak 15 3.2 21 4.5 32 6.9
Total 101 22.0 204 44.0 158 34.0
Perceptions of leaf-feeding insects and other arthropods. Almost three-fourths of
the respondents (72.5%) perceived that leaf-feeding insects would cause a yield loss.
About 16% thought that they would not cause a yield loss. About 43.5% of the respondents
perceived that leaf-feeding insects should be controlled early. Some 15%
believed that leaf-feeding insects need not be controlled, whereas 42% had no opinion.
Exactly two-thirds (66%) of the farmers also thought that leaf-feeding insects
feeding on plants in the early stage of growth can cause severe damage to the crop,
whereas 18% believed otherwise. Some 16% had no opinion.
Fewer than one-half of the respondents indicated that all arthropods in the field
would not cause damage to the rice plant. It is interesting to note that around 34%
admitted that they did not know the effect of arthropods on the rice plant, whereas
22% said that they cause damage. Table 6 shows responses from the different provinces.
Knowledge of natural enemies
Despite their perception that not all arthropods in their field would damage the rice
plant, many farmers were not aware of the presence of these arthropods in their fields.
According to 44% of the respondents, spiders, damselfly, dragonfly, preying mantis,
ground beetles, frogs, and wasps would not eat on the rice plants. Other orthopterans
(cricket and Conocephalus) and lady beetles were also mentioned (Table 7).
110 Rapusas et al

Table 7. Natural enemies mentioned by respondents. No farmer surveyed in Xieng Khouang mentioned a natural enemy.
Natural enemies Luang Oudumxay Luang Sayabouly Vientiane Vientiane Borikhamxay Savannakhet Saravane
Namtha Prabang Mun. Prov.
29 16
0 8
4 0
0 0
2 0
0 0
3 0
0 0
0 0
2 2
6 0
0 0
0 0
0 0
0 0
0 0
0 0
Champassak Total
5
0
1
0
0
0
0
0
0
0
1
0
2
2
1
7
1
126
36
21
11
20
6
13
2
12
5
10
1
2
2
3
12
4
Farmers mentioning (no.)
Spiders
Damselfly
Dragonfly
Preying mantis
Lady beetle
Ground beetle
Waterbug
Plantbug
Wasps
Conocephalus
Cricket
Water strider
Parasites
Predators
Birds
Frogs
Fish
10 16
2 12
0 6
5 6
1 1
0 1
0 1
1 1
0 12
0 0
0 0
0 0
0 0
0 0
1 0
0 2
0 3
13
4
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
2
1
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
19
2
5
0
5
1
1
0
0
0
2
0
0
0
1
3
0
9
1
5
0
4
0
1
0
0
0
0
1
0
0
0
0
0
7
6
0
0
7
4
6
0
0
0
0
0
0
0
0
0
0

Table 8. Yield comparisons of respondents who used and did not use pesticides.
Province
District
Sprayed
Unsprayed
Area Yield Area Yield
(ha) (t ha -1 ) (ha) (t ha -1 )
Northern region
Luang Namtha 2.5 2.56 29.6 2.96
Oudumxay
2.1 2.55
20.4 1.54
Luang Prabang 28.3 1.84
13.7 2.17
Sayabouly 4.5 1.96
33.0 1.61
Xieng Khouang 0 0
21.6 1.37
Central region
Vientiane Mun. Nasaythong 31.0 1.29 74.1 1.70
Saythany 14.4 2.66 70.5 1.93
Vientiane Prov.
19.8 2.63 25.0 3.00
Borikhamxay
4.8 3.20 22.5 2.77
Savannakhet Xaybouly 37.9 1.73 71.6 1.89
Champone 8.5 0.96 26.7 1.49
Southern region
Saravane 8.6 2.18 31.4 2.21
Champassak Champassak 7.6 1.48 26.3 1.93
Pakse 17.0 1.87 5.7 1.44
Sanasonboun 7.6 1.48 50.2 1.63
Total 197.0 480.7
Mean 2.05 2.11
Variance 0.855833 1.05666
Yield comparisons of respondents who applied
and did not apply insecticides
When average yields per hectare of respondents who applied and did not apply pesticides
were compared, the majority of those who did not apply pesticides reported
higher yields, although the differences were not significant (Table 8).
Knowledge on diseases
The respondents did not consider diseases to be a major constraint to rice production,
although some farmers reported observing blast, sheath blight, and false smut in their
fields.
Discussion
Rice production in Lao PDR has largely not been commercialized and is focused on
meeting home consumption needs. The level of adoption of improved production
technology is low, which reflects a combination of factors, including a lack of reliable
technical advice appropriate to conditions in the country. New, improved higher yielding
varieties are becoming available, and area-specific soil nutrient management practices
are being formulated. The national extension service is poorly developed and the
112 Rapusas et al

staff often lack the necessary technical expertise to adequately advise farmers. In
addition, even with access to technical advice and appropriate inputs, smallholders’
limited purchasing power can prevent them from adopting improved production practices.
But some changes are apparent with the development of a national research
program with support from IRRI and the Swiss Agency for Development and Cooperation.
For pest management, the survey showed that farmers are aware of the presence
of pests in their fields. More awareness and higher levels of technology adoption,
such as application of fertilizer and insecticide, had been noted near the main population
centers such as Vientiane. This reflects the concentration of extension and projecttype
assistance in these areas more than in others.
The majority of those who did not apply pesticides obtained higher yields, although
the differences were not significant. The insignificant differences seem to
indicate that applying pesticides for rice is not necessary. Two possible reasons for
the low use of pesticides are the lack of resources to buy the chemicals and the absence
of crop losses from pests, except isolated reports of severe crab and gall midge
damage. If farmers had more access to resources, pesticide use could increase. If this
happened, the existing natural ecological balance in the rice environment might be
affected markedly, which could lead to pest outbreaks. At this stage, we need to improve
farmers’ pest management decision making. Furthermore, the Pesticides Act to
control the import, manufacture, and repacking of pesticides, as well as their sale in
the country, must be enforced and implemented. Otherwise, pesticides that have been
restricted or banned in other countries will continue to find their way into local markets.
Most respondents were aware of the presence of other pests that do not feed on
rice in their fields such as spiders, dragonflies, crickets, and beetles, but were not
aware of the roles of these arthropods in the field. Educating farmers on the concept
of natural enemies in their fields should be a part of any training program on pest
management. Extension personnel should likewise be well versed in the concepts of
pest management to be able to provide farmers with the appropriate information and
technology. Doing so could lead to improvements in farmers’ well-being and in the
status of the farming community in the country.
References
Khotsimuang S, Schiller JM, Moody K. 1995. Weeds as a production constraint in the rainfed
lowland environment of the Lao PDR. In: Proceedings of the 15th Asian-Pacific Weed
Science Society Conference, 24-28 July 1995, Tsukuba, Japan, p 444-454.
Lao PDR. 1996. Basic statistics about the socio-economic development in the Lao PDR. State
Statistical Center, Ministry of Economic Planning and Finance, Vientiane, Lao PDR.
Lao-IRRI Project Report. 1996. Vientiane, Laos.
Lathvilayvong P, Schiller JM, Phommasack T, Kupkanchanakul T. 1995. Nutrient management
in the rainfed lowland rice environment of Laos. In: Proceedings of the International
Rice Research Conference, 13-17 February 1995, Manila, Philippines, p 267-278.
Pest management practices of rice farmers in Lao PDR 113

Lim GS, Heong KL. 1984. The role of insecticides in rice integrated pest management. In:
Judicious and efficient use of insecticides on rice. Manila (Philippines): International Rice
Research Institute. p 19-39.
Mumford JD, Norton GA. 1984. Economics of decision making in pest management. Annu.
Rev. Entomol. 29: 157- 174.
Norton GA, Heong KL. 1988. An approach to improving pest management in rice in Malaysia.
Crop Prot. 7:84-90.
UNDP (United Nations Development Programme). 1994. Development cooperation: Lao
People’s Democratic Republic. 1993 report.
Notes
Author’s addresses: H.R. Rapusas, Entomology and Plant Pathology Division, International
Rice Research Institute, Los Baños, Laguna, Philippines: J.M. Schiller, Lao-IRRI Project,
Vientiane, Lao PDR; V. Sengsoulivong, National Agricultural Research Center, Naphok,
Vientiane, Lao PDR.
Citation: Heong KL, Escalada MM, editors. 1997. Pest management of rice farmers in Asia.
Manila (Philippines): International Rice Research Institute.
114 Rapusas et al

CHAPTER 8
Pest management practices of rice
farmers in the Muda and Kemubu
irrigation schemes in peninsular
Malaysia
R. Normiyah and P.M. Chang
A dramatic change has occurred in rice cultivation in peninsular
Malaysia; direct seeding has now become the dominant crop establishment
method, replacing traditional planting. Surveys were made
on farmers’ pest management practices in the Muda and Kemubu
irrigation schemes in 1991 and 1994, respectively. Structured personal
interviews were used to obtain sociodemographic data together
with varietal use, seed rate, knowledge of pests and natural enemies,
pest management practices, including patterns of pesticide use, and
related safety measures. A comparison of the two survey areas
showed that although most farmers were able to recognize the major
insect pests, especially those found in their areas, and the major
grassy weeds, they could not identify diseases other than tungro.
They were aware of natural enemies but were uncertain about their
role. Most practiced early season insecticide application against leaf
feeders and chemical control appears to be their main management
tactic. They were not particularly concerned with associated health
hazards. Although they were aware of the role of clean and quality
seeds, many farmers could not obtain them because of limited supply.
Based on the studies, some proposals are made to improve
farmers’ pest management practices in both areas.
Introduction
The scenario of rice cultivation in Malaysia has changed dramatically over the past 25
years (1970-94) since rice double cropping began. The rapid adoption of nitrogenresponsive
modern rice varieties and the provision of irrigation facilities have led to
modifications in the microclimate in ricefields, resulting in changes in species dominance
of rice pests. The widespread transformation from manual transplanting to direct
seeding because of labor shortages (Ho and Md. Zuki 1988) has aggravated the
competitive interactions of flora and fauna in ricefields. The adoption of direct seeding
gained momentum in the rice agroecosystem in the 1980s and it is now the pre-
115

dominant form of rice crop culture in Malaysia. The closed canopy and increased
plant density predispose the rice crop to higher risks of insect and disease pathogens
(Chin 1985). In addition, with direct seeding, grassy weeds and rice seeds germinate
simultaneously, thus allowing grassy weeds to flourish and compete with rice.
Several programs have been implemented to improve pest management in rice
cultivation, and many research projects have been conducted under the banner of
integrated pest management (IPM), wherein different control tactics were deployed.
In many cases, farmers did not continue these IPM practices after the projects ended
because they perceived them to be financially nonbeneficial or too tedious. This could
be because no significant yield increase was observed after practicing IPM, although
chemical use may be reduced. Moreover, the need to go to ricefields for frequent
surveillance under IPM was a burden, especially to old farmers.
Most pest management research focuses on understanding biological and ecological
processes that occur in the agroecosystem, with less attention to socioeconomic
and implementation aspects (Heong et al 1995b). The problem of implementation
should be viewed as an integral part of research and development aimed at designing
appropriate pest management practices. Researchers need to understand why
farmers carry out certain management practices.
In Malaysia, research on socioeconomic constraints to pest management adoption
is still lacking, although there is increasing realization of such a need (Heong
1984, Heong et al 1985, 1987, Normiyah et al 1995). Although biological and ecological
studies of pests are essential to understand and improve pest management
strategies, there is a need to consider related socioeconomic aspects. Heong et al
(1994) noted that obtaining such background information to identify the key features
of real pest management problems may be the most important factor in determining
research and extension priorities.
The status of pest management practices of rice farmers was studied in the Muda
irrigation scheme in 1991 (Normiyah et al 1995) and in the Kemubu irrigation scheme
in 1994. This paper examines and compares the changes in pest management practices
of rice farmers for these two areas. Agronomic practices, knowledge on pest
management, pesticide use, and safety measures are discussed.
Survey sites
The Muda irrigation scheme is located on the coastal alluvial plain in the northwest
corner of peninsular Malaysia (Fig. 1). Straddling the two northern states of Kedah
and Perlis, the area has traditionally been a net exporter of rice within the country and
is popularly known as the Rice Bowl of Malaysia. It is the largest of eight designated
rice granary areas in peninsular Malaysia and accounts for about 45% of the nation’s
rice production. The scheme covers about 96,000 ha of riceland and is operated by
63,000 farm families. About 47% of these are owner-operators, 25% are tenants, 17%
are owner-tenants, and 9% are others, each managing a farm of about 2 ha (MARDI
1984). The widespread transformation of the rice cultivation technique from manual
transplanting to direct seeding in the 1980s has resulted in extensive changes in the
116 Normiyah and Chang

Fig. 1. Rice granary areas in peninsular Malaysia and locations of the
study: (A) Muda irrigation scheme, (B) Kemubu irrigation scheme.
traditional rice habitat. The most conspicuous change is the shift of weed dominance
from less competitive broadleaf weeds and sedges to more competitive grasses (Ho et
al 1990a, b, Azmi and Baki 1995). Nik Mohd. Noor and Hirao (1987) reported that
ricehoppers were the most important insect pests in the Muda area and the incidence
of insect pests was higher in direct-seeded fields than in transplanted fields.
The Kemubu irrigation scheme is located on the northeastern plain of the state of
Kelantan (Fig. 1). It is the second largest of the eight rice granary areas. The scheme
covers about 31,450 ha of ricefields and is operated by 45,000 farm families. About
47.5% of these are owner-operators, 11% are tenants, 25.5% are owner-tenants, and
19% are others. The average farm size is less than 1 ha. The widespread change in rice
Pest management practices of rice farmers in peninsular Malaysia 117

cultivation technique from transplanting to direct seeding during the 1990s, which
occurred later than in Muda, has also created a significant impact on the pest spectrum
in the Kemubu area, particularly in the weed dominance from broadleaf weeds
and sedges to grasses (Azmi and Mashhor 1995). There is no obvious change in the
composition of insect pests; rodents remain a serious threat to rice production.
Of these two areas, Muda adopted direct-seeding culture rapidly (from the early
1980s) compared with Kemubu, where direct seeding only started to gain ground
from 1990 onward. To date, more than 95% of the Muda area is under direct seeding,
versus 50% of the Kemubu area. Chemical use, especially herbicides, by Muda farmers
is more prevalent than with Kemubu farmers. The type of chemicals used was also
found to be different. This gives us a good opportunity to study the similarities and
differences between the two areas in their rice pest management practices, other than
chemical use, and perhaps make suitable recommendations.
Methods
To obtain data on rice farmers’ pest management practices, structured personal interviews
were used with a questionnaire designed for the purpose. The questions were
pretested on rice farmers from the two areas and subsequently modified to avoid ambiguities.
The survey was conducted in March 1991 in the Muda area, where a sample of
250 rice farmers was selected randomly for the study. Similarly, 200 rice farmers
from Kemubu were interviewed in May 1994. The farmers were interviewed by a
team of trained enumerators from the Department of Agriculture (DOA) and the Malaysian
Agricultural Research and Development Institute (MARDI).
The data obtained were analyzed and frequency tables were generated based on
the number of farmers who responded rather than the total sample. In cases where
multiple responses were obtained, the sample size was used. The general linear models
procedure of SAS (SAS 1985) was used for the analyses of variance and the F test
was used to determine association between variables.
Survey results
Sociodemographic profile
Respondents from the Muda area were mainly Malays (98%), Chinese (1.3%), and
Siamese (0.8%), whereas the Kemubu farmers were all Malays (Normiyah et al 1995).
About 52% of the Muda farmers and 76% of the Kemubu farmers were more than 50
yr old. Of the respondents from both areas, 78% in Muda and 96% in Kemubu were
more than 40 yr old.
The formal educational background of the farmers varied in the two schemes. In
Muda, more than two-thirds of the farmers had a primary education (1–6 yr of schooling),
16% completed secondary school (7–11 yr of schooling), 8% attended religious
school, and 12% had not attended school at all (Normiyah et al 1995). In the Kemubu
118 Normiyah and Chang

area, 66% of the respondents had only a primary education and 34% had no formal
education.
Agronomic practices
MR 84 was the most popular rice variety planted by farmers in the two areas. In
Muda. another variety, IR42, was also widely planted. but not in Kemubu. Instead,
other MR varieties (those recommended by MARDI) such as MR 106, MR 103, and
MR 127 were preferred. MR series cultivars were also planted by farmers in Muda.
Evidently, rice farmers at both locations planted more than one cultivar.
Healthy, clean, and weed-free seeds are important, especially in direct-seeding
cultivation. Farmers obtained rice seeds from several sources. Some 48% of the respondents
from Muda and 59% from Kemubu bought seeds from the seed production
center of the DOA, which is the sole supplier of healthy clean seeds at both sites.
About one-third of the farmers used seeds from their own stock (seeds harvested from
the previous season), whereas the rest of the farmers obtained or exchanged seeds
with friends and neighbors.
The recommended seeding rate for direct seeding is 60 kg ha -1 (MARDI 1984).
The survey revealed that 96% of the Muda farmers who practiced direct seeding used
more seeds than the recommended rate as they believed that the higher seed rate
increased crop yield and helped suppress weeds. This compares with the situation in
Kemubu, where two-thirds of the respondents used the recommended rate and only
19% used higher rates.
Knowledge of pests and natural enemies
The majority of farmers from these areas could recognize rice pests (Table 1). More
than 60% of the farmers in both areas recognized rice pests such as rice bug (Leptocorisa
spp.), Malayan black bugs (Scotinophara), leaf feeders, and stem borers. Muda farmers
could recognize brown planthoppers (BPH), whereas farmers in Kemubu did not
even mention this pest. Few farmers knew about whitebacked planthopper (WBPH)
and green leafhopper (GLH). The majority of the Muda farmers knew about the symptoms
of tungro attack but they seemed less knowledgeable about other rice diseases.
Similarly, the Kemubu farmers did not mention much about rice diseases. But most
respondents were able to recognize the prevailing weeds such as Echinochloa crusgalli,
Marsilea crenata, Leptochloa chinensis, sedges, Sphenoclea zeylanica, and
Ischaemum rugosum.
Respondents were also asked to rank pests according to their perception of pest
destructiveness to the rice crop. From the responses, it was apparent that rodents,
insects, and weeds were considered to be the three most destructive pests. Rodents
were mentioned as the most destructive pest, the one that caused the highest crop loss,
followed by insects (particularly BPH attack in Muda), weeds (mainly Echinochloa
spp.), and diseases. Rodents were reported as the main pest by 41% of the Muda
farmers, whereas leaf feeders were mentioned as the main pest by 58% of the Kemubu
farmers. Farmers from both regions reported weeds, particularly E. crus-galli.
Pest management practices of rice farmers in peninsular Malaysia 119

Table 1. Farmers’ pest and natural enemies identification skills.
Rice pests
Respondents a (%)
Muda
Kemubu
Insect pests
Brown planthopper 74.1 11.5
Whitebacked planthopper 37.1 0.9
Green leafhopper 41.8 2.7
Malayan black bug 66.7 90.3
Leaf feeder 67.7 97.3
Stem borer 68.5 77.0
Rice bug 72.9 96.5
Diseases
Tungro
Blast
Bacterial disease
75.7 27.4
23.9 1.8
14.3 -
Rice weeds
Echinochloa crus-galli 90.0 90.7
Leptochloa chinensis 65.7 30.1
Ischaemum rugosum 36.7 36.3
Sphenoclea zeylanica 47.8 54.0
Marsilea crenata 73.3 65.5
Sedges 59.8 77.0
Natural enemies
58.0 38.0
a Multiple responses.
About 58% of the respondents from Muda were aware of the prevailing natural
enemies in rice, whereas 38% of those from Kemubu knew about them (Table 1). The
natural enemy species known included spiders, frogs, dragonflies, and barn owls;
some mentioned beetles.
Pest management practices
On average, more than 90% of the farmers from both areas reported using chemicals
every season to control rice pests. Virtually all the rice farmers (90%) who practiced
direct seeding in Muda adopted chemical weed control, a practice being picked up by
Kemubu farmers (50%) because of their more recent adoption of direct seeding and
subsequent grassy-weed problems.
Most rice farmers tend to apply their first insecticide spray during the first 40 d
after crop establishment. Many respondents (47% from Muda and 52% from Kemubu)
reported their first insecticide application in the first 30 d after sowing. Their main
target was leaf feeders.
Farmers sampled from the two areas were divided into two groups based on early
and late spraying of insecticides. To determine the effect of early and late spraying on
yield, an analysis of variance was computed. For each group, mean yield and standard
deviation (SD) were obtained and the F test was used to test for significance.
120 Normiyah and Chang

For Muda farmers, a mean yield of 5,070 kg ha -1 and SD of 716 for early sprayers
and a mean yield of 5,280 kg ha -l and SD of 313 for late sprayers were obtained.
These results show no significant difference in the yields of farmers who spray early
and late (F = 1.91, P = 0.17). The yield variation for those who spray early was greater
compared with those who spray late.
For farmers who spray early in Kemubu, mean yield was 3,458 kg ha -1 with SD
of 878, whereas a mean yield of 3,304 kg ha -1 and SD of 857 were found for farmers
who spray late. Similarly, the results show no significant difference in the yields of
farmers who spray early and late in Kemubu (F = 0.77, P = 0.38). The yield variation
for the two groups was about the same. This means that farmers who sprayed insecticide
early did not necessarily obtain higher yields than those who sprayed late (after
the first 30–40 d after sowing).
Pesticide use
Table 2 shows trends in pesticide use by respondents from the two sites. Muda farmers
commonly used buprofezin, an insect growth regulator, effective against hoppers
and safe for natural enemies and fish (Nik Mohd. Noor et al 1989). Other insecticides
used were MIPC, BPMC, buprofezin + MIPC, and MTMC + fenthoate to combat
hoppers. Respondents also reported using BPMC to control Scotinophara other rice
bugs, and leaf feeders. Few farmers mentioned fenthion, which is also effective in
controlling rice bugs. Some farmers used carbofuran to control stem borers.
Kemubu farmers used insecticides such as endosulfan, alphacypermethrin, BPMC,
carbofuran, fenthion, and methamidophos. A variety of insecticides were used to control
leaf feeders. Endosulfan and alphacypermethrin appeared to be widely used. Farm-
Table 2. Types of pesticides used to control major pests.
Commonly used chemicals
Major pests
Muda
Kemubu
Echinochloa spp.
Molinate + 2,4-D butyl ester
Fenoxaprop-ethyl
Propanil-based herbicides
Quinclorac
Pretilachlor + safener
Bensulfuron
Hoppers
BPMC, buprofezin Endosulfan
(BPH, WBPH, GLH) a Buprofezin + MIPC
MTMC + fenthoate
Rice bugs BPMC Endosulfan
Alphacypermethrln
BPMC
Methamidophos
Stem borer
Carbofuran
Endosulfan
Carbofuran
Leaf feeders BPMC Endosulfan
Alphacypermethrln
Rodents Zinc phosphide Zinc phosphide
Chronic rodenticides
Chronic rodenticides
a BPH = brown planthopper, WBPH = whitebacked planthopper, GLH = green leafhopper,
Pest management practices of rice farmers in peninsular Malaysia 121

ers in both areas did not mention the use of chemicals to control diseases. In fact, it
was found that some farmers used insecticides to treat rice diseases.
More than 90% of the Muda farmers and about 50% of the Kemubu farmers used
herbicides to control weeds. Phenoxy herbicides such as 2.1-D butyl ester and 2,4-D
amine were mainly used to control broadleaf weeds and sedges. Muda farmers used
molinate and propanil-based herbicides to control barnyard grass (Echinochloa spp.),
whereas Kemubu farmers used quinclorac, fenoxaprop-ethyl, and bensulfuron.
Most farmers used acute rodenticides (zinc phosphide) to control rodents and
only a few farmers reported using chronic rodenticides.
Safety measures
About 47% of the Muda farmers and 13% of the Kemubu farmers kept pesticides in
safe places, whereas many admitted negligence in that they kept and placed the chemicals
anywhere, such as in rice storage sheds, under the house, and even in toilets. The
results also showed that most farmers did not follow adequate safety precautions during
spraying (Normiyah 1995). In addition, it was found that some farmers (44%
from Muda and 26% from Kemubu) burned or buried used pesticide containers, contrary
to recommendations. They usually discarded the used pesticide containers in
ricefields, in irrigation canals, near streams, around the house, into unused wells, and
into the surrounding undergrowth.
Discussion
The introduction of double cropping has changed the rice production scenario in both
the Muda and Kemubu irrigation schemes, thus contributing to the increase in cropping
intensity. But the most dramatic change has been the widespread transformation
from manual transplanting to direct seeding during the 1980s in Muda and during the
1990s in Kemubu. This has made a significant impact on the pest spectrum, particularly
the appearance of more competitive grasses and the higher incidence of insect
pests and diseases in direct-seeded fields, probably because of modifications in the
crop microenvironment.
Associated with this is a peculiar type of rice weed in Malaysia that farmers refer
to as weedy rice or padi angin. This group of volunteer rice cultivars has different
agronomic characteristics, with the common ones of easy grain shattering. It is highly
undesirable because it reduces yield (Abdul Wahab and Suhaimi 1991). It is now an
important research priority because of its potential threat to rice production in Malaysia
if it is not contained in time.
Varieties
Although the surveys show that farmers use a number of rice varieties, farmers distinctly
prefer one — MR 84 — which has been popular since the mid-1980s in Malaysia.
Although MARDI has released new varieties, farmers still prefer MR 84 and may
take time to adopt the newer ones.
122 Normiyah and Chang

Use of clean seeds is a basic component of weed control: this is highly important
from the viewpoint of disease control and prevention of the spread of weedy rice.
Although many farmers are aware of the usefulness of clean and high-quality seeds,
their availability and supply remain a problem. There are no private seed producers in
Malaysia and DOA is the sole organization that provides clean and high-quality seeds.
But its capacity is limited, and it normally supplies only about 30% of the total seed
requirements. Therefore, many farmers keep a portion of their harvest for seeds or
buy them from their neighbors.
A seeding rate of 60 kg ha -1 should be sufficient for direct seeding (MARDI
1984). At higher rates, weeds may be suppressed to a certain extent without increasing
rice yields (Azmi and Mashhor 1990). So when weeds are controlled, higher
seeding rates are not justified. Unfortunately, many farmers, especially in Muda, are
still seeding at more than 100 kg ha -1 . These wasteful high seeding rates have aggravated
the shortage of clean seeds and increased input cost. In cases where insecticide
or fungicide applications are required toward the end of the vegetative stage, high
seed rates result in dense canopies, which makes application difficult especially if the
target pest is below the canopy.
Identification of pests and natural enemies
Farmers involved in the study had showed some ability to recognize rice pests. Twothirds
of the respondents recognized insect pests such as rice bugs, black bugs, leaf
feeders, and stem borers, whereas few could recognize GLH and WBPH. Muda farmers
could recognize BPH because of their experience with outbreaks, whereas Kemubu
farmers were not familiar with the pest because of the absence of outbreaks there.
Kemubu farmers did recognize leaf-feeder damage and were concerned about this
pest.
Diseases were not mentioned much by the farmen in the two areas, which suggests
that extension efforts may not have focused on rice diseases yet, other than
tungro. Diseases such as sheath blight and bacteria leaf blight are common in directseeded
fields and red stripe is becoming more important. Other bacterial diseases and
grain discoloration are also becoming common.
Most farmers could identify the most noxious weeds, such as E. crus-galli, and
the other major grassy weeds. Now farmers must be more alert to the growing importance
of weedy rice, which can cause a considerable yield loss (Abdul Wahab and
Suhaimi 1991).
Respondents also knew about the natural enemies of rice pests but were not very
sure of their role in pest management. More training courses and extension efforts
will have to be directed at improving farmers’ pest identification skills and their understanding
of the role of natural enemies.
Pesticide use
Farmers were found to rely on pesticides as the major means of pest control. Herbicides,
insecticides. and rodenticides were mainly used.
Pest management practices of rice farmers in peninsular Malaysia 123

In direct seeding, manual weeding is difficult and not practical because of the
labor shortage. Chemical weeding is the most popular method among rice growers in
both Muda and Kemubu as it has been found to be labor saving and effective. The
different choice of herbicides between the two areas probably reflects the amount of
promotion by different chemical companies (Table 2).
The data show that farmers practiced early season insecticide application. Their
main target pests were leaf feeders. Insecticide use by farmers during the early crop
season seems to be based on perceived needs and perhaps fear, rather than real needs,
because of the high visual impact. Early season insecticide use does not benefit rice
production. It can be detrimental to the ecological balance and cause secondary BPH
problems (Chang 1992, Heong et al 1995a). Field infestation with as much as 67%
damaged leaves was reported to cause no yield loss (Miyashita 1985, Rubia et al
1995). Leaf feeders that damage rice plants in the early stages do not cause enough
loss to justify spraying. Thus, early sprays against leaf feeders are not economically
justifiable. In addition, spraying of broad-spectrum insecticides such as
alphacypermethrin by Kemubu farmers will reduce the natural enemy population
(Chang 1992). Thus, we need to get farmers to reduce early insecticide applications.
A new approach, farmer participatory research (FPR), complementary to extension
activity, is a useful method for encouraging farmers to engage in small-scale research
(learning by doing) together with researchers and extension agents so that they can
adapt new technology and also extend it to other farmers (Escalada and Heong 1993).
In Malaysia, the use of broad-spectrum insecticides is not recommended and
MADA (Muda Agricultural Development Authority) has taken positive steps, following
MARDI’s recommendations, to encourage rice farmers to use selective insecticides
such as buprofezin with low toxicity to mammals, fish, and natural enemies.
The acceptance of buprofezin has been encouraging. Its use, either alone or in combination
with MIPC, helps contain BPH and prevent it from escalating.
Kemubu farmers used a variety of insecticides against leaf feeders. Endosulfan
and alphacypermethrin were widely used. The choice of these two insecticides indicated
the tendency of farmers to spray insecticides that have a rapid effect on target
pests. Availability of chemicals also influenced the choice of insecticides, as 84% of
the farmers stated that they purchased their insecticides from the farmers’ association,
which had a large stock of these chemicals. Kemubu farmers thus appear to lack the
proper knowledge on the choice of chemicals, particularly insecticides.
Endosulfan is hazardous not only to farmers and applicators but also to nontarget
organisms, particularly fish. Being an organochlorine, it is quite persistent in the soil
and groundwater.
The surveys indicated very little use of chemicals against diseases, which reflects
the degree of recognition of diseases by rice farmers in both areas.
Farmers from both regions also need to be discouraged from their preference for
acute rodenticides. Not only are many chronic poisons available, but other technologies,
such as traps and barrier systems (Lam et al 1987) and the use of barn owls,
should be integrated into a comprehensive rat control program.
124 Normiyah and Chang

Direct seeding need not increase insect pest outbreaks. More importantly. farmers’
practices, especially early season insecticide applications, can upset natural control
mechanisms and allow more frequent BPH population buildups. The excessive
use of herbicides to clear bund vegetation during land preparation can also destroy
refuge areas for natural enemies of rice insect pests (Chang 1992).
Farmers were usually aware of the health hazards caused by pesticide use, but
they did not treat this with concern. Many farmers did not keep chemicals in safe
places and neither did they bum or bury used chemical containers. Inadequate storage
facilities, lack of use of protective clothing during spraying, unsafe handling, and
inefficient sprayer maintenance all contribute to a more hazardous environment, not
only to farmers, but to farming communities as well. This is another important aspect
that has been overlooked in pest management practices, by both farmers and authorities,
as pointed out by Rola and Pingali (1993). This attitude toward safety measures
in pesticide handling must be changed and improved, and health and medical considerations
should be emphasized more in the future (Rola and Pingali 1993).
Conclusions
This study has showed the different levels of understanding and practices in pest management
of rice farmers in Muda and Kemubu. Although this is related to differences
in the time of adopting direct seeding, other factors such as farmers’ education, extension
systems, and social differences need to be considered.
Farmers from Kemubu usually lacked knowledge on the choice of chemicals,
particularly insecticides. On the contrary, farmers in Muda were more informed and
followed more sound pest management practices. Farmers were also influenced by
their experiences with the dominant pest species in each locality. As such, extension
programs need to be tailored to target these differences in order to be more effective.
In Kemubu, greater attention should be paid to changing farmers’ perceptions of leaffeeder
damage and to improving the choice of chemicals before serious ecological
disruption occurs. Thus, our main task in rice leaf-feeder management appears to be
related more to seeking changes in current farmer practices (Heong 1992).
Emphasis should also go toward recognizing management practices for rice diseases,
especially sheath blight, since this can be a severe yield constraint in directseeded
rice.
We also propose that studies be made on the long-term effect of herbicide use on
nontarget organisms, as this may tip the ecological balance in favor of pest species.
For both regions, emphasis must continue on extension to equip farmers with the
necessary pest management knowledge to enable them to make on-farm decisions.
This would help ensure the adoption of the IPM philosophy at the farm level, and
contribute significantly to sustainability and environmental health as well as to the
well-being of the farming community.
Pest management practices of rice farmers in peninsular Malaysia 125

References
Abdul Wahab AH, Suhaimi O. 1991. Padi angin: ciri-ciri, kemudaratan dan cara-cara
memhasmikannya. (Padi angin: characteristics, adverse effects and methods of its eradication).
Teknologi Padi, Malaysian Agricultural Research and Development Institute 7:27-
31.
Azmi M, Mashhor M. 1990. Competition of barnyard grass (E. crus-galli (L.) Beauv.) in seeded
rice. Proceedings of the 3rd International Conference on Plant Protection in the Tropics,
20-23 Mar 1990, Genting Highlands, Malaysia. p 224-229.
Azmi M, Baki B. 1995. The succession of noxious weeds in tropical Asian rice fields with
emphasis on Malayan rice ecosystems. Proceedings of the 15th Asian Pacific Weed Science
Conference, 24-28 Jul 1995, Tsukuba, Japan.
Azmi M, Mashhor M. 1995. Weed succession from transplanting to direct-seeding method in
Kemubu rice area. Malaysia J. Biosci. 6:2.
Chang PM. 1992. Rice leaffolder and brown planthopper occurrence in FELCRA, Seberang
Perak, Malaysia. Proceedings of the lnternational Workshop on Economic Threshold Level
for Rice Leaffolder in China, 4-6 Mar 1992, Beijing, China. p 14-15.
Chin KM. 1985. Reports of new sheath diseases of rice in Peninsular Malaysia. Teknologi Padi,
Malaysian Agricultural Research and Development Institute 2:79-83.
Escalada MM, Heong KL. 1993. Communication and implementation of change in crop protection.
In: Chadwick DJ, Marsh J, editors. Crop protection on sustainable agriculture.
Chichester: John Wiley & Sons. p 191-202.
Heong KL. 1984 . Pest control practices of rice farmers in Tanjong Karang, Malaysia. Insect
Sci. Applic. 5:221-226.
Heong KL. 1992. Rice leaffolders: are they serious pests? Proceedings of the International
Workshop on Economic Threshold Level of Rice Leaffolders in China, 4-6 Mar 1992,
Beijing, China. p 8-11.
Heong KL, Ho NK, Jegathesan S. 1985. The perception and management of pests among rice
farmers in the Muda Irrigation Scheme, Malaysia. Report No. 105. Kuala Lumpur (Malaysia):
Malaysian Agricultural Research and Development Institute. 11 p.
Heong KL, Yahya H. Tee SP. 1987. Pest management practices of tobacco farmers in Bachok,
Kelantan, Malaysia. Report No. 11 6. Kuala Lumpur (Malaysia): Malaysian Agricultural
Research and Development Institute. 9 p.
Heong KL, Escalada MM, Lazaro AA. 1995a. Misuse of pesticides among rice farmers in
Leyte, Philippines. In: Pingali PL, Rogers PA, editors. Impact of pesticides on farmers’
health and the rice environment. Kluwer Academic Publishers. p 97-107.
Heong KL, Escalada MM, Vo Mai. 1994. An analysis of insecticide use in rice: case studies in
the Philippines and Vietnam. Int. J. Pest Manage. 40(2): 173- 178.
Heong KL, Teng PS, Moody K. 1995b. Managing rice pests with less chemicals. GeoJournal
35(3):337-349.
Ho NK, Asna Booty O, Rabirah A. 1990a. Herbicide usage and associated incidences of poisoning
in the Muda area, Malaysia. Proceedings of the 3rd Tropical Weed Science Conference,
4-6 Dec 1990, Kuala Lumpur, Malaysia. p 321-333.
Ho NK, Md. Zuki I. 1988. Weed population changes from transplanted to direct seeded rice in
the Muda area. Proceedings of the National Seminar and Workshop on Rice Field Weed
Management, 7-9 Jun 1988, Penang, Malaysia.
126 Normiyah and Chang

Ho NK, Md. Zuki I, Asna Booty O. 1990b. The implementation of strategic extension campaign
on integrated weed management in the Muda area. Proceedings of the 3rd International
Conference on Crop Protection in the Tropics, 20-23 Mar 1990, Genting Highlands,
Malaysia. 20 p.
Lam YM, Mohd. Salleh AA, Lee AK. 1987. A trapping device for rice field rat control. Teknologi
Padi, Malaysian Agricultural Research and Development Institute 3:29-32.
MARDI (Malaysian Agricultural Research and Development Institute). 1984. Panduan amali
keperluan dan syor-syor untuk padi tabur terus dan tanaman padi cara cara mengubah.
Serdang, Malaysia: MARDI. 21 p.
Miyashita T. 1985. Estimation of economic injury in the rice leaf roller, Cnaphalocrosis medinalis
Guenee (Lepidoptera: Pyrilidae): relations between yield loss and injury of rice leaves at
heading or in the grain filling period. Jpn. Appl. Zool. 29:73-76.
Nik Mohd, Noor NMS, Habibudin H, Chang PM, Ahmad Puat N, Mohamed MS. Abdul Latif
AZ. 1989. Buprofezin, an insect growth regulator for the control of ricehoppers. Teknologi
Padi, Malaysian Agricultural Research and Development Institute 5:1-6.
Nik Mohd, Noor NMS, Hirao J. 1987. Status of rice pests in direct-seeded fields in the Muda
area. Teknologi Pradi, Malaysian Agricultural Research and Development Institute 3:39-
44.
Normiyah R, Chang PM, Azmi M, Aznan A. 1995. Pest management practices of rice farmers
in the Muda Irrigation Scheme, Malaysia. Makalah Sesekala Bil. 14. Kuala Lumpur (Malaysia):
Malaysian Agricultural Research Development Institute. 14 p.
Rola AC, Pingali PL. 1993. Pesticides, rice productivity, and farmers’ health: an economic
assessment, Los Baños (Philippines): International Rice Research Institute.
Rubia EG, Shepard BM, Yambao EB, Ingram KT, Arida GS, Penning de Vries F. 1995. Stem
borer damage and grain yield of flooded rice. J. Plant Prot. Trop. 6(3):205-211.
SAS (Statistical Analysis System). 1985. User’s guide to statistics. Version 5 edition. Cary
(NC, USA): SAS Institute, Inc.
Notes
Authors’ addresses: R. Normiyah, Malaysian Agricultural Research and Development Institute
(MARDI) Research Station. P.O. Box 105, 05710 Alor Setar, West Malaysia; P.M. Chang,
MARDI Research Centre, P.O. Box 203. Pejabat Pos Kepala Batas, 13200 Seberang Perai,
West Malaysia.
Acknowledgments: The authors would like to express their appreciation to the Swiss Agency
for Development and Cooperation (SDC), which supported the surveys through the IPM
Network coordinated at IRRI. Special thanks go to Dr. K.L. Heong, of IRRI’s Entomology
and Plant Pathology Division, for making these studies possible. The authors also
wish to thank Harun Hamid, research assistant at MARDI, Alor Setar, for his assistance in
data processing; and all the enumerators from MARDI and DOA, for their help in conducting
the survey.
Citation: Heong KL, Escalada MM, editors. 1997. Pest management of rice farmers in Asia.
Manila (Philippines): International Rice Research Institute.
Pest management practices of rice farmers in peninsular Malaysia 127

CHAPTER 9
Farmers’ perceptions of rice tungro
disease in the Philippines
H. Warburton, F.L. Palis, and S. Villareal
We conducted a survey in the Philippines to investigate farmers’
perceptions of rice tungro disease. Farmers regarded tungro as a
serious problem even in areas where outbreaks are rare. Most farmers
could describe tungro disease symptoms but gaps existed in
their understanding of the causes and modes of spread. The relationship
between the virus disease and its main vector, green leafhopper
Nephotettix virescens (Distant), was often not clearly understood.
Many thought that tungro could be transmitted through water,
air, soil, and other insects. Farmers were unaware of the risks of
leaving infected plants, which can act as sources of disease inoculum,
in the field.
Farmers’ pest management strategies were based on the use
of insecticides and resistant varieties. In areas where tungro was
common, farmers had greater awareness of the importance of resistant
varieties.
Factors affecting perceptions of tungro were analyzed, including
links between beliefs about human diseases and plant diseases.
The risk characteristics of tungro were also examined to explain why
farmers regard tungro as such an important disease.
This study highlights the need for more relevant information for
farmers that would complement their current knowledge and enable
them to improve their management of this disease.
Introduction
Over the years, farmers have developed methods of managing plant diseases. Many
examples exist where farmers’ agronomic practices have provided an adequate and
environmentally sustainable means of combating diseases (Thurston 1990). Changes
and intensification within the farming system, however, can bring increased disease
pressures. In particular, plant diseases that occur sporadically but affect large areas
pose difficulties for farmers because of their unpredictability and risk of major crop
loss. Local knowledge and practices must evolve to cope with these problems.
129

Farmers’ knowledge of their local ecosystem is often extensive, based on observations
and experimentation in the field (Chambers et al 1989). But farmers’ perceptions
of plant diseases may differ considerably from those of scientists. Farmers’ local
knowledge may be substantial in some aspects, but incomplete in others. Bentley
(1992) noted that a farmer’s depth of knowledge about pests is related to the importance
and visibility of the pest,
Because disease pathogens are not easily seen, it is often difficult for farmers to
account for the causes of diseases. Beliefs about plant diseases may also be related to
beliefs about human diseases because notions of health and sickness are common to
both (Fairhead 1991). Knowledge systems within a farming community are dynamic;
new observations and information from within and outside the community are constantly
being incorporated into existing beliefs, so perceptions of plant diseases are
likely to change over time.
If scientists are to work with farmers in improving management options for plant
diseases, they need to understand what farmers know. This is especially relevant for
introducing integrated pest management (IPM). IPM requires sufficient knowledge
of pests and diseases for farmers to formulate appropriate management strategies that
may include a mix of cultural, biological, and chemical methods. Acquiring sufficient
skills to implement this knowledge-intensive system is often a constraint to adopting
IPM (Goodell 1984, Matteson 1992, Bentley and Andrews 1996).
We carried out a study in the Philippines on farmers’ perceptions of rice tungro
disease. This disease is associated mainly with intensively cultivated rice areas and
tends to occur in sudden, sporadic outbreaks. Such epidemics can affect large areas
and result in heavy crop losses. The objectives of the study were to investigate and
document farmers’ knowledge of the disease symptoms and causes. determine their
management practices, and assess the factors affecting their perceptions.
Rice tungro disease in the Philippines
Rice tungro disease occurs in South and Southeast Asia and can result in severe crop
losses (Ou 1985). Tungro is a virus disease that cause’s stunting of the rice plant and
yellow to red discoloration of the leaves. If infected during early growth stages, the
rice plant will not produce panicles; at later stages, there may be panicles but with low
grain fill. Tungro is transmitted by several species of leathopper, the most important
of which is the green leafhopper Nephotettix virescens (Distant).
Major epidemics of tungro were recorded in the Philippines in the 1970s. The
Department of Agriculture (DA) issued recommendations on control and organized
mass spraying campaigns against the green leafhopper. Since then, several hotspot
areas have been recorded where tungro is endemic, and sporadic outbreaks have occurred
elsewhere.
Recommendations for tungro control have changed over the years: in the 1970s,
attention was focused mainly on controlling the insect vector with insecticides (BPI
1971, DA 1979). More recent recommendations have emphasized nonchemical control
methods (DA 1994) because of a concern about pest resistance and resurgence
130 Warburton et al

problems caused by insecticides—as well as concerns about the health and environmental
risks these chemicals pose. Researchers have also recognized that chemicals
may not be very effective in controlling tungro anyway. Current control strategies
propose a three-way approach to managing tungro based on
• using disease-resistant varieties;
• removing sources of disease inoculum through roguing, plowing under infected
crops, removing stubble, planting early, and scheduling for fallow periods;
and
• controlling the vector.
Methods
Data were gathered through a series of focus group interviews and individual interviews
of farmers using a questionnaire. The focus group interviews were undertaken
first to gain an understanding of the issues surrounding tungro and to explore reasons
behind farmers' perceptions and practices. The farmer survey was used to follow up
on issues that emerged from the group discussions and to include farmers with different
backgrounds so that the variability in perceptions and practices among individuals
could be assessed.
Interviews were conducted in different parts of the country chosen to include
areas with differing incidence of tungro and intensity of rice production. These included
tungro hotspots (Polangui in Albay, Santo Tomas, Davao del Norte; Bansalan,
Davao del Sur), areas with recent outbreaks (Famy, Laguna), and other irrigated and
rainfed areas where tungro occurred less frequently (Santa Arcadia in Cabanatuan
City; Lupao and Muñoz, Nueva Ecija). Ten group discussions were held in six villages
during October and November 1994, with an average of six men and four women
in a group. Farmers were contacted through the local DA extension technicians.
Two hundred forty-two farmers, randomly chosen from the four locations. were
interviewed using the structured questionnaire between August and December 1995.
Farmers were first asked about insect pests and diseases in general—rather than
only about tungro—to gauge the relative importance of this disease without undue
bias toward it. Later, farmers were asked specifically about tungro. Emphasis was
placed on using local names and descriptions rather than scientific terms, and farmers
were not prompted with any mentions of the tungro vector, green leafhopper.
Results
Rice was the major crop in all the villages. Intensity of rice production varied from
one crop a year in rainfed areas to three crops a year where water was available yearround.
Farm sizes ranged from 0.1 to 9 ha. averaging 1.8 ha. Crop rotation was rarely
practiced outside the rainfed areas.
All farmers had heard of tungro, and the majority said it had occurred in their
fields. Only 13% said they had never experienced it, whereas 6% reported experiencing
an outbreak of tungro more than five times. Except for the hotspot areas, farmers
Farmers' perceptions of rice tungro disease in the Philippines 131

ecalled that most of the outbreaks had occurred in the 1970s or 1990s with far fewer
occurrences during the 1980s.
Importance of tungro
The farmers considered tungro to be an extremely damaging disease, especially if it
occurred during the early crop stages. If that happened, most farmers believed that the
infected plants would not yield at all. If tungro struck at later stages, they said they
would harvest a little, but crop losses of more than 70% were reported. Few farmers
thought that rice plants could recover after being infected. One group described tungro
to be like the disease AIDS, while another group said it was like cancer.
Farmers ranked tungro as the most serious pest problem in both the wet and dry
seasons in all the areas surveyed, including those where tungro outbreaks are infrequent.
Golden snail ( Pomacea sp. ) was ranked a close second (Table 1). Farmers
explained that they ranked tungro highly among pests-despite the relatively low
frequency of outbreaks—because it was so damaging when it occurred and so difficult
to control.
Knowledge of tungro symptoms, causes, and mode of spread
Farmers' descriptions of the disease symptoms usually corresponded with those of
scientists, but wide discrepancies occurred over perceptions of the causes and mode
of spread (Table 2). Examples of incompatible descriptions are leaves turn white,
worms cause tungro, and tungro “germs” are in the soil.
Table 1. Farmers’ ranking a of insect pests and diseases.
Dry season
Wet season
Tungro (268)
Tungro (374)
Golden snail (247) Golden snail (279)
Rats (172) Rats (204)
Rice bug (80) Stem borer (87)
Stem borer (75) Rice bug (75)
Brown planthopper (45) Brown planthopper (62)
Birds (34) Green leafhopper (44)
Green leafhopper (20) Birds (23)
Other insect pests and diseases (

Table 3. Farmers’ perceptions of causes and spread of tungro a .
Farmers’ perceptions % of 242
respondents
Caused by
Susceptible plant variety
Soil problem (too acidic,
lack of fertilizer, too much fertilizer)
Weather conditions
Seeds
Green leafhopper
Insects
Disease present in soil
Virus
Water
Brown planthopper
Worms in plants
Late planting time
Other
Do not know
Spread by
Flying insects
Water
Air
Seeds
Worms/virus in soil
Roots of infected plants
Stem borers
Other
Do not know
22.3
17.8
17.8
16.1
14.0
11.1
6.6
6.2
5.8
2.5
2.1
2.1
2.9
12.0
37.6
14.5
10.3
7.0
3.7
1.2
0.8
2.1
30.6
a Multiple answers possible.
Recognizing symptoms. The majority of farmers could describe the symptoms of
tungro: yellow or red leaves, stunting, low tillering, and low yield. Only 12% of the
respondents described symptoms not associated with tungro, such as molds or folded,
cut, or white leaves. In group discussions, farmers supplied additional detailed information
they use to identify tungro from other conditions, such as its occurrence in
patches in the field, or the fact that rice does not recover after fertilizer is applied if it
is infected with tungro. Most farmers (79%) could also name other diseases or conditions
of rice with symptoms similar to those of tungro, such as yellowing of leaves.
Lack of fertilizer was the most commonly reported condition.
Causes and spread of tungro. Opinions differed over the causes of tungro and
how it spread (Table 3). The majority thought that tungro occurrence was connected
with the variety or seeds used and with insects, but the causal mechanism was unclear.
Another vague area centered around what farmers who mentioned virus actually
imagined a virus to be. Some described it as a “micro-worm.”
Farmers, particularly in areas where tungro occurs frequently, regarded a rice
variety’s susceptibility to the disease as a major causal factor. For example, 53% of
the respondents in tungro hotspot Polangui thought that variety or seeds were impor-
Farmers’ perceptions of rice tungro disease in the Philippines 133

tant causal factors, whereas only 16% of the respondents in the rainfed areas mentioned
them.
The survey results revealed the importance many farmers place on soil condition
as a determining factor. Weather conditions, particularly sudden changes or excessive
heat, waterlogged fields, and even the volcanic eruption of Mount Pinatubo were
cited as conditions associated with tungro incidence.
Almost a third (31%) of the respondents said they did not know how tungro
spreads. Although 38% were aware that insects feeding on infected plants spread
tungro, even those farmers who had attended DA seminars did not perceive clearly
the green leafhopper’s exact role as the major vector. Several farmers thought that
there were modes of disease transmission in addition to the green leafhopper. Some
also regarded green leafhopper as a pest in its own right. independent of its function
as a tungro vector.
Accurate identification of the green leafhopper is sometimes uncertain because
local names are not always detailed enough to be unique. In Central Luzon, for example,
the Tagalog word ngusong kabayo is usually translated as green leafhopper,
but it may also mean, generically, adult hopper in some locations. In Bicol, the general
word layog-layog, meaning flying insects, is commonly used.
Many farmers thought that tungro could be spread through water, soil, or the air,
with some describing it as a germ that is carried in the air like the human cold virus.
Many farmers from the tungro hotspots were convinced that the disease is seed-borne:
20% of the respondents from Polangui, for example, thought this. Farmers in Polangui
and Davao experienced a major tungro outbreak after the DA introduced new seeds
from outside the area. This occurrence reinforced their conviction that tungro is in the
seeds.
Source of disease inoculum. Farmers appeared unaware of the threat to new rice
plantings from diseased plants in the surrounding area. No one mentioned that the
virus remains in the plant, providing a source of inoculum for green leafhoppers to
spread to other plants. Even farmers who knew that tungro is a virus and is spread by
green leafhoppers believed that the insects would always carry the virus, meaning it is
a permanent, intrinsic characteristic of the insect. They were not aware that a green
leafhopper must feed initially on an infected plant to acquire the virus, and that an
insect will lose its infectivity if it does not repeatedly feed on diseased rice plants.
Additionally, they did not know that only a portion of a green leafhopper population
may carry the virus.
Management strategies
Farmers’ management of tungro depends on when the disease is noticed in the field
(Table 4). Chemical control and changing the variety were the most widely used strategies.
About a third of the farmers, however, said that they did not take any measures
against tungro, especially if it occurred during late crop growth. No control measure
was thought effective enough to be worth the cost or effort.
Chemical control. Many farmers (77%) said that they would use insecticides to
control tungro. Farmers believed that these chemicals were effective in killing the
134 Warburton et al

Table 4. Farmers’ tungro management strategies.
% of 242 farmers using the measure
Control measure When tungro At early crop At late crop To prevent
occurs in growth growth reoccurrence
nearby field
next season
Use insecticide
(spray/granules)
Change variety
Rogue infected plants
Plow under field
Drain or control water
Apply fertilizer
Burn stubble
Practice fallow
Use early or
synchronized planting
Clean field and dikes
Use traditional repellents
Use other controls
Do nothing
Do not know
70.2
6.6
1.7
0
14.0
3.7
0.4
0
0.8
3.7
0.8
3.7
12.4
1.2
72.7
1.2
39.3
16.5
14.5
8.3
0.4
3.0
0.4
0.4
0
4.1
5.8
1.7
53.7
1.2
5.4
2.1
9.5
6.2
0.4
0
0
0
0
1.2
31.8
1.2
11.6
85.5
0.4
3.7
5.4
5.4
8.7
11.0
3.3
0
0
5.8
3.3
0
insects that cause tungro (and simultaneously protecting against other insect pests),
but they were aware that spraying did not effectively control the spread of tungro.
Commonly used insecticides were monocrotophos, endosulfan, and cypermethrin.
Some farmers used furadan granules to treat seedbeds. Only one respondent thought
that fungicides were effective.
Use of resistant varieties. Most farmers (86%) said that they would change varieties
to prevent tungro from occurring again. Awareness of the importance of varietal
resistance to tungro was most acute in the hotspots. Farmers in these areas said that
resistance to tungro was one of the main criteria for selecting seed. In the group discussions,
they listed the varieties grown locally and specified which were susceptible
and resistant. Many actively sought and tested new lines for resistance. Farmers knew
that some varieties that had initially appeared resistant had become infected over
time, meaning the resistance had broken down. In certain cases, farmers reported
varieties as susceptible to tungro, although the DA still regarded them as resistant.
It is unclear exactly what farmers mean by “resistant.” Farmers outside the endemic
areas did not always differentiate clearly among resistance to various insect
pests and diseases. Some believed that newly released varieties were always more
resistant than older ones. Other farmers believed that it was not good to plant the same
seed on the same land for consecutive seasons, so they obtain new seed even though it
might be of the same variety. (They say that the soil has become tired (sawa) of the
same seed.) Sometimes the variety being used could not be identified because it had
been given a local name.
Farmers' perceptions of rice tungro disease in the Philippines 135

Removing diseased plants. Roguing diseased plants was regarded as laborious
and ineffective. It was only thought appropriate if a few plants were affected at early
crop stages. None of the farmers indicated any urgency in removing diseased plants.
Some farmers said that they would plow under the whole field if it became infected.
Others stated that plowing under is not a feasible option, especially when the
crop is near harvest: they would rather wait and reap whatever they can. Plowing a
field is an expensive option, and farmers, because the harvest is lost, may lack the
funds to pay for it. Plowing under stubble immediately after harvest was not practiced
as a measure against tungro but was done by those growing three crops a year who
wanted to plant another crop as quickly as possible.
Planting dates, fallows, and crop rotations. Adjusting planting dates or changing
cropping patterns to incorporate fallows or break crops were not major components
of farmers’ management strategies. The most awareness of planting date effects was
among farmers in Famy, where many who had planted late in 1994 had experienced
tungro. Additionally, the DA had been advising farmers to allow for a fallow period
after outbreaks.
Many factors unconnected with tungro influenced farmers’ decisions about early
or synchronous planting, including the availability of water, money, and labor to start
preparing land. Synchronous planting and fallows are difficult to implement-specially
in areas where water is continuously available. Farmers in those places can
plant when they like, and implementing a fallow may mean having to forgo a third
crop. The farmers surveyed did not mention crop rotation as a way to manage tungro;
those who practiced it did so in response to water shortages.
Other control methods. Several farmers cited draining the field as a way to manage
tungro—specially in Polangui, where respondents reported tungro to be more
prevalent in waterlogged fields. Farmers drain fields to control other pests-especially
snails—so it is not clear whether this control is always aimed specifically at
tungro. Several farmers also reported applying fertilizer as a way to reduce losses
from tungro.
Before pesticides were common, farmers said that they used certain plants
(kakawate, Gliricida sepium, and tagbag or talbak, Garcinia linearifolia) to ward off
diseases. Branches were placed in the inlets of canals going into a field, or sometimes
at the corner of a field. Only two of the farmers interviewed were using these methods.
Differences in knowledge and practices among farmers
Results from the farmer survey were analyzed to investigate whether any patterns
between farmers’ knowledge and practices relating to tungro and socioeconomic factors
emerged. Chi-square tests did not reveal any significant patterns between the
number of tungro occurrences experienced, age of farmer (as a proxy for farm experience),
or level of formal education and knowledge of the symptoms, causes, and
mode of spread. No difference was evident between men and women. Farmers in
locations where tungro was endemic did not understand its mode of spread more than
those in other areas (in terms of compatibility with scientists’ knowledge).
136 Warburton et al

Differences in knowledge between locations, however, were evident from the
group discussions. Farmers had more detailed knowledge of the susceptibility of different
varieties to tungro and were more skeptical about the effectiveness of chemicals
in areas where tungro is common.
Information sources on tungro
About half (49%) the farmers reported DA technicians to be the most common source
of advice on tungro, but the recommendations received on its control varied from area
to area. Learning from other farmers (20%) and from their own experience (13%)
were also important. Only 6% said that they obtained information from pesticide
sellers. Some farmers were skeptical about advice from chemical companies. Several,
for example, suggested that pesticide companies deliberately put “germs” into
their products to ensure that farmers continue to buy more pesticides to control them.
According to the farmers, most of the DA technicians’ recommendations included
using either spray or granular insecticides. Roguing, plowing a whole field, fertilizing,
and draining a field were also commonly advised. These recommendations, however,
were not always followed: some farmers reported believing that it was too expensive
to spray as advised or to plow under a whole field.
Some farmers also received advice from DA technicians on suitable varieties, but
details about which varieties to use varied. In some areas, for example, DA technicians
advised farmers to avoid planting IRRI varieties. Except in Famy, farmers received
little advice on planting dates and fallow periods.
Discussion
Identifying tungro and its vector
Interviews alone cannot determine with certainty farmers’ abilities to identify tungro,
but the majority of the descriptions given were consistent with scientific views. Farmers
do not, however, seem to use “tungro” as a general name for yellowing or stunting.
Misidentifying green leafhopper and its role is more prevalent among farmers
than misidentifying tungro. Some do not distinguish clearly between insect types,
while others know of green leafhopper but do not understand the relationship between
the insect and the disease. Those who think that green leafhopper is a pest in its
own right (apart from being a vector of tungro) are not in line with current scientific
opinion that the insect is unlikely to cause significant damage if it is not infective.
Farmers may think that all types of hoppers are pests—as are brown planthoppers in
large numbers—or that old extension messages from the DA and current messages
from chemical companies have raised the status of green leafhopper as a pest in farmers’
minds. (Insecticide labels often mention green leafhopper control.)
Results from this study correspond with those from a study of farmers’ perceptions
of tungro in Malaysia, which revealed that farmers could describe tungro symptoms
but were unclear about the causal agent (Heong and Ho 1987). But the importance
attached to using resistant varieties in the Philippines was not evident in the
Malaysian survey: only 5% of 270 farmers reported using them. A possible reason for
Farmers’ perceptions of rice tungro disease in the Philippines 137

the difference may be that Philippine farmers in endemic areas have had more years’
experience dealing with the disease—and observing the performance of varieties—
than had the Malaysian farmers in 1984.
Farmers’ perceptions of the cause of disease
Based on Bentley’s typology of farmer knowledge (Bentley 1992), tungro disease is
likely to fall into the category of important but difficult-to-observe, where farmers’
beliefs may differ considerably from those of scientists and possibly involve more
folklore or rituals. Results from our study are in line with this typology because farmers
do find it difficult to explain the causes of tungro, and their accounts differ from
scientific views. Remedies for tungro based on folklore, however, did not play a major
part in farmers’ strategies.
The parallels between perceptions of human and plant diseases are supported by
the information gleaned through this study. In Tagalog, the same word ( sakit ) is used
for plant and human diseases. Pesticides are often referred to as medicine ( gamot ) or
sometimes poison ( lason ). Farmers often likened tungro to the human illnesses of
AIDS and cancer. The belief that tungro is spread by air, water, and soil as well as
insects is understandable if farmers assume that tungro, because it is a disease, will
spread in ways similar to those of human germs.
Studies in the Philippines of beliefs about the causes of human diseases have
revealed that people often have complex views of the cause and effect of disease, with
several causes often associated with one or more effects, rather than one cause leading
to one effect. Often a combination of a person’s susceptibility (perhaps caused by
hunger or fatigue) plus an unexpected natural phenomenon (such as a sudden heavy
rain) is thought to result in sickness (Himes 1971).
Throughout the Philippines, numerous drug company advertisements ensure that
the population is “germ” conscious. The term “virus” is well known, with antiseptics
and other products commonly used against these unseen organisms. Tan (1987), however,
notes that although theories of germs are recognized, they may not always be
well understood.
Although farmers find it difficult to describe exactly what causes tungro and
some express the view that it is the will of God ( bahala na ), they are not fatalistic or
passive in their approach to looking for control methods. Several causes or conditions
based on the plant’s state and on natural phenomena are often given to explain the
outbreak of tungro. Based on an approach similar to that of Himes (1971) and Tan
(1987), the perceived causes of tungro can be understood as a mixture of things that
make the plant more susceptible to infection occurring simultaneously with natural
conditions likely to cause tungro (Fig. 1). Farmers’ strategies to combat tungro seem
to be based on these perceptions because they often attempt to strengthen the plant’s
condition and reduce the causal agents.
What is missing from this scheme is an understanding of the role of diseased
plants in providing a source of inoculum for the vectors. Without this knowledge, it is
not surprising that farmers pay little attention to diseased plants remaining in the
field.
138 Warburton et al

Fig. 1. Farmers' perceptions of causes and conditions leading to tungro
infection, and controls for the disease.
Risk perceptions of tungro
The probability of an outbreak of tungro, compared with pests such as golden snails,
is small even in endemic areas. Despite this, farmers rate tungro as an important pest
even in nonendemic areas. Another study of perceptions of insect pests and diseases
(Heong et al 1992) also revealed the importance farmers accord to tungro.
Farmers’ perceptions of the characteristics of risk from a tungro outbreak are
very different from those of snails. Farmers regard tungro as an uncontrollable disease
from which a plant cannot recover. They also believe that the occurrence of
tungro is difficult to predict and that its cause cannot be seen. In contrast. snails appear
to be easier to observe and control, and rice can be replanted (Fig. 2). People
tend to perceive risks with unknown and uncontrollable aspects as greater than those
with known and more controllable characteristics—even though the probability of
the event is much less (Slovic 1987). Farmers’ perceived powerlessness in predicting
or controlling tungro may explain why they rate it so highly.
Effects of farmers' perceptions on adopting management strategies
Farmers know that tungro is connected with insects and rice varieties, so they have a
rationale for adopting strategies for insect control and experimenting with different
varieties. In some cases, farmers may be ahead of scientists in their knowledge of a
variety’s usefulness and local conditions. IR60, a tungro-resistant variety, provides an
excellent example. The Philippine Seed Board officially released it as a variety—but
only after farmers in Mindanao had acquired, tested, and disseminated the breeding
line IR13429-299-2-1-3 themselves (Fujisaka 1990). Further investigation of the varieties
farmers are testing may yield information that researchers could incorporate
into location-specific varietal trials.
Farmers’ perceptions of rice tungro disease in the Philippines 139

Fig. 2. Characteristics of farmers’ risk perceptions of insect
pests and diseases. Adapted from Slovic 1987, Fig. 1, for
general risks.
Management strategies connected with removing diseased plants (roguing, plowing
under a field, practicing fallows) are more difficult to implement because farmers
do not fully understand the effect of these on tungro. Until they realize the potential
problems from leaving diseased plants in the field, they have no incentive to remove
them. Extension messages need to focus on this information gap.
Conclusions
Farmers’ knowledge of tungro is uneven. By simply observing the disease in their
fields, farmers have developed some effective management strategies—such as selecting
resistant varieties—without detailed knowledge of tungro’s causes. Yet misunderstandings
over how the disease is transmitted mean that farmers put their rice
crops at risk by leaving diseased plants in fields.
Farmers’ perceptions of insect pests and diseases are important influences on
their adoption of control methods. They may not require knowledge of every scientific
detail, but they need to understand the reasons for adopting control measures,
especially when these measures require time and effort. When recommending strategies
for pest management, scientists must provide relevant information to farmers
that will build on what they already know. Filling these gaps to improve understanding
will enable farmers to make more informed decisions about management strategies.
References
Bentley JW. 1992. The epistemology of plant protection: Honduran campesino knowledge of
pests and natural enemies. In: Gibson RW, Sweetrnore A, editors. Proceedings of a seminar
on crop protection for resource-poor farmers. Chatham (UK): Natural Resources Institute
and Technical Centre for Agriculture and Rural Co-operation. p 107-118.
Bentley J, Andrews K. 1996. Through the roadblocks: IPM and Central American smallholders.
Gatekeeper Series No. 56. London: International Institute for Environment and Development.
20 p.
140 Warburton et al

BPI (Bureau of Plant Industry). 1971. Tungro: its symptoms and control. Manila (Philippines):
BPI.
Chambers R, Pacey A, Thrupp LA. 1989. Farmer first: farmer innovation and agricultural research.
London: Intermediate Technology Publications.
DA (Department of Agriculture). 1979. Masagana 99 emergency bulletin: easy and sure means
of controlling tungro. Manila (Philippines): DA.
DA (Department of Agriculture). 1994. Instituting preventative and eradication measures against
the tungro virus in rice. Memorandum Circular No. 3, 30 June 1994. Manila (Philippines):
DA.
Fairhead J. 1991. Methodological notes on exploring indigenous knowledge and management
of crop health. In: RRA notes: participatory methods for learning and analysis. No. 14.
London: International Institute for Environment and Development.
Fujisaka S. 1990. Farmer knowledge and sustainability in rice farming systems: blending science
and indigenous innovation. IRRI Social Science Division Paper No. 90-39. Los Baños
(Philippines): International Rice Research Institute.
Goodell G. 1984. Challenges to international pest management research and extension in the
Third World: do we really want IPM to work? Bulletin of the ESA. Fall 1984. p 18-26.
Heong KL, Escalada MM, Lazaro AA. 1992. Pest management practices of rice farmers in
Leyte, Philippines. Los Baños (Philippines): International Rice Research Institute. 57 p.
Heong KL, Ho NK. 1987. Farmers’ perceptions of the rice tungro virus problem in the Muda
irrigation scheme, Malaysia. In: Tait J, Napompeth B, editors. Management of pests and
pesticides. Boulder (Colorado, USA): Westview Press. p 165-173.
Himes RS. 1971. Tagalog concepts of causality: disease. In: Lynch F, de Guzman A, editors.
Modernisation: its impact in the Philippines. Vol. V. Institute of Philippine Culture Paper
No. 10. Quezon City (Philippines): Ateneo de Manila University.
Matteson PC. 1992. ‘Farmer first’ for establishing IPM. Bull. Entomol. Res. 82:293-296.
Ou SH. 1985. Rice diseases. Wallingford (UK): CAB.
Slovic P. 1987. Perception of risk. Science 236:280-288.
Tan ML. 1987. Usug, kulam, pasma: traditional concepts of health and illness in the Philippines.
In: Traditional medicines in the Philippines: Research Report No. 3. Quezon City
(Philippines): Alay Kapwa Kilusang Pangkalusugan.
Thurston HD. 1990. Plant disease management practices of traditional farmers. Plant Dis.
74(2):96-102.
Notes
Authors’ addresses: H. Warburton, Natural Resources Institute (NRI), Central Avenue, Chatham
Maritime, Kent ME4 4TB, United Kingdom; F.L. Palis and S. Villareal, International Rice
Research Institute (IRRI), P.O. Box 933. 1099 Manila. Philippines.
Acknowledgments: We express our thanks to the staff members of the Department of Agriculture
at Santo Tomas, Digos, Polangui, Famy, Lupao, and Muñoz who assisted us during
the survey. We are grateful to Drs. K.L. Heong, Paul S. Teng, and Prabu L. Pingali at IRRI.
and Dr. S.R. Obien and staff members at the Philippine Rice Research Institute for their
support. Many thanks are due to Dr. Tim Chancellor and the NRI-IRRI Tungro Project
team members, and to Mr. Joseph Addawe and Ms. Maria Aurea Princena for processing
data. Most of all, we thank all the farmers who participated in the study.
Citation: Heong KL, Escalada MM, editors. 1997. Pest management of rice farmers in Asia.
Manila (Philippines): International Rice Research Institute.
Farmers’ perceptions of rice tungro disease in the Philippines 141

CHAPTER 10
Weed management practices
of rice farmers in Iloilo, Philippines
K. Moody, M.M. Escalada, and K.L. Heong
To understand rice farmers’ perceptions and practices in weed management,
we conducted a survey among 300 rice farmers in lloilo
Province, Philippines. In both the wet and dry seasons, IR64 was
planted by the majority of rice farmers, using seeds obtained from
other farmers and neighbors, their own seed stock, the Department
of Agriculture, and private seed growers. Farmers who retained seeds
for planting in the next season used seeds from the second crop
ping as these were judged to have better quality. Most farmers
changed seeds every year in order to plant a disease-resistant variety,
to improve yield, or to try other varieties. Off-types, weedy types,
and red rice were the contaminants observed in the rice crop. When
contaminants were present, farmers either removed them or did
nothing. Rice weeds were believed to have come from the soil, to be
carried by irrigation water, or to be mixed with rice seeds. Among the
weeds considered to cause production losses, Echinochloa crusgall;
L. was reported to be the most destructive. Other weeds considered
destructive were E. colona (L.) Link, Cynodon dactylon (L.)
Pers., Fimbristylis miliacea (L.) Vahl, and Cyperus iria L.
To reduce weed problems, herbicide applications, water management,
and hand weeding were the most common measures practiced
by farmers. Herbicides were reportedly applied by most farmers
in the seedling and vegetative stages. In using water management
to control weeds, most farmers either flooded their fields a
few days after applying herbicide or flooded the ricefield without using
herbicides. Water management was the main responsibility of
the male in the household, whereas weeding was done mostly by
hired labor.
In terms of attitudes toward seed management, most farmers
believed that seeds from private seed growers did not require additional
cleaning, that infection and contaminants in rice seeds for
planting decreased yield, and that neighbors and friends should exchange
seeds among themselves. Farmers were ambivalent, however,
about the value of seed cleaning and the function of winnowing
in removing all infected seeds.
143

Introduction
Agricultural production losses have been directly attributed to weeds. Estimates of
the annual losses caused by weeds have been placed at 109, or an equivalent of 46
million t (Mortimer 1994, Moody 1991). On most farms, inadequate weeding has
resulted in serious yield reductions (Parker and Fryer 1975). With changes in crop
establishment to direct seeding and increasing labor and water management costs,
weed problems have become increasingly important (Moody 1993). Not only has the
shift from transplanting to direct seeding brought about a change in weed distribution,
it has also resulted in greater weed competition because hand weeding has become
more difficult (Estorninos and Moody 1982, Ho 1990).
Weeds can be controlled through the application of diverse methods—manual,
mechanical, cultural, and chemical. Among these options, cultural practices remain
an integral part of any weed control program. The simultaneous application of sound
crop management practices and herbicide use has been noted to bring about effective
weed control (Day 1972). In the Muda area of Malaysia. the decline in herbicide use
during 1990-93 was attributed to a strategic extension campaign on integrated weed
management. Farmers who used proper cultural practices and correct timing and dosage
of herbicides achieved better weed control than those with poor land preparation
and improper water management (Adhikarya 1994).
To control weeds, proper cleaning procedures are also necessary to ensure crop
seeds with high purity as planting material. The use of clean seeds minimizes the
spread of weeds in the field, which reduces the cost of weed control, maintains crop
quality, reduces hosts of plant pests, decreases irrigation costs, reduces injury to livestock,
and increases land value (Doll 1994).
Although the emergence of weeds has been traced to various sources, many important
weeds in rice are introduced either through impurities in the rice seeds
(Delouche 1988, Fujisaka 1993) or from weed seeds submerged in the soil (Heong et
al 1995). At crop maturity, weed seeds are harvested and processed along with rice
seeds, thus contaminating the planting materials to be used for the next crop. Consequently,
weeds get widely distributed through rice seeds with mixtures of weed seeds
(Delouche 1988). Contamination of crop seeds with weed seeds can perpetuate weeds
in fields (Edward 1963).
Mortimer (1994) described the persistence of seed banks in the soil and the presence
of extremely high populations of underground meristems from which new plants
can grow into adults. Tuber populations of Cyperus rotundus L. can reach 10,000,000
ha -1 (Rao 1968), whereas the annual production of Imperata cylindrica can reach 6 t
ha -1 (Soerjani 1970). In experiments conducted in Malaysia, Watanabe and Azmi Man
(1995) reported millions of weed seeds per unit area, highlighting the importance of
seed banks in weed management.
The use of herbicides has been found to be a more economical and effective
means of weed control (Moody 1994a, Caseley 1994). The extent of herbicide use is
often influenced by the prevailing labor costs, rice prices, and other socioeconomic
and institutional considerations. Where labor is scarce and rice production is higher,
144 Moody et al

herbicides have been reported to be a viable substitute for labor (Heong et al 1995,
Denning 1983). Traditionally, farmers hand weed, but with the availability of inexpensive
herbicides, farmers are “pulled” toward becoming herbicide-dependent. For
many farmers, herbicides appear to be a more economical means of weed control than
hand weeding, resulting in greater reliance on herbicides. But weed scientists have
cautioned against excessive dependence on herbicides by small farmers and emphasized
the need to reduce this dependence among farmers (Alstrom 1990, Heong et al
1995, Moody 1993). At issue are the residual effects of herbicides in the soil, health
hazards to farmers, the increase in unemployment, and undesirable effects on the
environment (Labrada et al 1994, Akobundu 1987, Moody 1993).
Work by anthropologists at IRRI suggested that, unlike with insect management,
farmers are more knowledgeable and innovative in ways to manage weeds. But the
availability of cheap herbicides may have diminished the importance of some of the
innovative methods. Even in herbicide use, rice farmers had demonstrated vaned application
strategies. For instance, in Nueva Ecija, Philippines, Fajardo and Moody
(1990) reported that farmers used herbicide combinations, sequential herbicide applications,
nonconventional application methods, varying application times, and different
water management strategies.
Hand weeding, although generally regarded as drudgery, has been considered a
vital practice even where herbicides are used, to prevent increases in resistant or tolerant
weeds. Even when weed infestation is low, hand weeding is recommended to
hinder the development of weed seeds (Labrada et al 1994).
Despite dramatic increases in herbicide use in rice, crop losses caused by weeds
have not declined and may even have increased. Weed problems persist because of
their tremendous reproductive capacity and recycling potential, inadequate prevention,
ecological shifts, and changing weed populations (Heong et al 1995).
Farmers perceive and respond to insects, diseases, and weeds differently. In contrast
to the damage done by insects, snails, rodents, and plant diseases, farmers tend
to overlook lost productivity from weeds (Doll 1994). Because weeds are so common,
farmers in many areas may not fully realize the losses they cause and the cost of
control (Heong et al 1995). As a result, farmers fail to recognize the need for adequate
weed management.
To be able to design appropriate strategies to improve farmers’ weed control practices,
it is necessary to understand their perceptions and practices in weed management.
On-farm surveys and focus group interviews are aimed at a better understanding
of farmers’ needs and constraints, knowledge. attitudes, and practices. Information
on farmers’ needs can then be transformed into researchable problems and results
communicated back to farmers. Through this process, we can ensure that efforts are
directed at problem-solving.
In this study, we focused on rice farmers’ perceptions and practices in weed management.
The area selected was Iloilo, considered to be the top rice-producing province
in the Philippines, where direct seeding is the predominant method of rice establishment.
Weed management practices of rice farmers in Iloilo, Philippines 145

lloilo Province
As the second largest province in western Visayas, Iloilo occupies the southern portion
of Panay Island. Its borders are defined by Capiz Province to the north, Antique
to the west, Panay Gulf and Iloilo Strait to the southeast. and the Visayan Sea and
Guimaras Island to the east. The province has a total land area of 466,340 ha, 243,640
of which are cultivated with rice.
The topography of Iloilo Province varies from flatlands and rolling hills to mountain
peaks and ranges. The mountain ranges lie along the border between Iloilo, Antique,
and Capiz and roll down into a flat plain toward the coastal towns. Almost onethird
of the province is considered flat. The province has 17 soil types. The soil is
usually fertile and suitable to almost all types of agricultural crops. The loam soil is
predominant and conducive to farming.
About 150 rivers and creeks traverse the entire province and these have been
identified as possible sources of irrigation water. The Jalaur River basin, considered
as the major source of irrigation water, records the highest annual flow.
The climate in Iloilo has been classified into two types. Type 1 is along the southern
and central part, which has two pronounced seasons. dry from December to May
and wet from June to November. Maximum rain is from June to September. Type 2
can be found along the northern part of Iloilo, where the rainfall is evenly distributed
throughout the year.
Methods
To gain some insights into rice farmers’ weed management practices, we made preliminary
visits to the survey areas in coordination with agricultural and provincial
government officials. During farm visits, we conducted interviews with key informants
such as the municipal agricultural officers, private seed growers, and other
farmers.
Questionnaire development and pretest
After a series of meetings, a multidisciplinary team, composed of a weed scientist, an
economist, an entomologist, a plant pathologist, and a communication scientist, developed
a 92-item survey questionnaire on famers’ perceptions and practices in weed
management. The prototype questionnaire was pretested twice to determine farmers’
reactions to it, to validate the translation of key technical terms used, to find out
whether the technical terms could be understood, to ascertain whether the sequence
of the questions could solicit the desired information, and to estimate the length of
time required to complete the interview.
Using the draft questionnaire that had been translated into Hiligaynon, the language
spoken by the majority in Iloilo, the pretest involved individual interviews and
small group discussions with rice farmers and Department of Agriculture technicians.
The pretest generated useful feedback that suggested ways in which the questionnaire
could be improved.
146 Moody et al

Sampling procedure
Multistage sampling was used to randomly select 300 rice farmers from Iloilo Province.
The first stage involved a listing of rice-growing municipalities in the province.
Next, 10 municipalities were selected randomly from this list and, from each municipality,
three villages were selected randomly. Finally, 30 rice farmers were drawn
randomly from each village. To ensure objectivity, none of the survey respondents
had attended season-long IPM training. The list of sample villages was reviewed by
the municipal agricultural officers (MAOs) to determine whether a Farmers’ Field
School (FFS) had been conducted in those areas. Two municipalities where an FFS
had been conducted recently were replaced. Figure 1 shows the location of the survey
areas. Nine of the ten areas had 30 farmers sampled, whereas San Dionisio had 33.
Fig. 1. Location map of survey sites in lloilo Province, Philippines.
Weed management practices of rice farmers in Iloilo, Philippines 147

Data analysis
The survey data collected were coded and entered into a spreadsheet program using a
microcomputer. Frequency tables were generated using the FREQ procedure available
in Statistical Analysis System (SAS 1985). Percentages were based on the number
of farmers who responded rather than the total sample. In cases where multiple
responses were obtained, either the sample size or the number of farmers who responded
for a particular item was used.
Results
Sociodemographic profile of farmers
Of the 303 farmers interviewed, 72.3% were men and 27.7% were women. Almost
half (49.2%) of the respondents were lessees, 29% were owner-operators, and 16.8%
were tenants. Four farmers were cultivating rice areas that were used as loan collateral.
In terms of education, 49.5% of the respondents had an elementary education
and 27.1% had completed high school. Some 14.2% went as far as college and 5.3%
attended vocational school. Only 3.9% had no schooling.
Information sources
For farming information, 55.4% of the farmers interviewed resorted to the extension
technician, 31.4% indicated that they relied on themselves. and 21.8% turned to their
neighbors and friends. Other farm information sources mentioned included seminars
and schools (14.2%), the mass media (11.9%), and relatives (6.9%).
In terms of pest management advice, 75.9% of the farmers consulted the agricultural
technician, whereas 17.8% sought assistance from their friends and neighbors.
Fewer than 10% relied on their own experience and 3.7% turned to family members.
A few (2%) read instructions on pesticide labels and leaflets and 1.3% sought help
from their landowner.
Seed management
Varieties planted. Although farmers reported a wide range of rice varieties, the majority
planted IR64 for both the first crop (79.2%) and second crop (63.6%). Other farmers
grew IR36, IR60, and PSBRC10. The popularity of IR64 among Iloilo farmers
reflects the results of a recent seed management survey in Central Luzon (Diaz 1995).
Sources of rice seeds. Farmers were asked where they obtained their rice seeds in
1993, the year before the survey. Major sources mentioned by Iloilo farmers were
exchanging seeds with other farmers and neighbors (38.9%), their own seed stock
(32%), the Grains Production Enhancement Program (GPEP) of the Department of
Agriculture, private seed growers (15.2%), and the farmers’ cooperative.
Seed selection. To have the best seeds, farmers chose seeds from a particular
cropping season. According to farmers, seeds from the first crop were often used to
pay land rent (8.1%) and had been found to have good quality (24.9%), although
other farmers (20.7%) likewise judged the seeds from the second crop to have good
quality as well.
148 Moody et al

As in the survey in Central Luzon (Diaz 1995), many of the Iloilo farmers (62.2%)
we interviewed changed their seed every year, and 30.4% did so every 2–3 yr. A few
changed their seed once in 4–5 yr (0.6%) and the same number replaced their seed
after more than 5 yr (0.6%). Farmers changed seeds in their desire to plant a diseaseresistant
variety (31.7%), to increase yield (22.8%), and to simply try other varieties
(24.58).
Seed contaminants. Farmers reported observing contaminants in their rice crop.
Among these, off-types were specified by 67.3% of the farmers, followed by weedy
types (53.8%) and red rice (49.2%). When farmers found contaminants, they removed
these mainly by hand roguing off-types (52.4%), weedy types (49.1%), and red rice
(51.48). Other methods used included changing seeds, sorting, soaking seeds in water,
and drying seeds. Some 43.4% of the farmers interviewed did nothing to contaminants
(Table 1).
Weeds in ricefields
Farmers reported perceiving several rice weeds in their ricefields. The most commonly
mentioned was Fimbristylis miliacea (L.) Vahl, followed by Echinochloa crusgalli
(L.) P. Beauv, and Paspalum conjugatum Berg (Table 2). These were also the
Table 1. Farmers’ responses to contaminants found in the rice crop,
Iloilo, dry season 1995.
Reponses Off-types Weedy types Red rice
(multiple) (n=204) (n=163) (n=149)
(%)
Changed seeds 8.3 3.7 6.4
Rogued contaminants 52.4 49.1 51.4
Sorted seeds 7.8 3.7 2.1
Soaked seeds in water 0.9 1.2
0.0
Dried seeds 0.0
12.3
0.0
Applied herbicide 0.0
2.4
2.1
Did nothing
34.8 58.9 36.4
Table 2. Weeds reported by farmers to be present in ricefields,
Iloilo, dry season 1995.
Weed
Farmers
No. %
Fimbristylis miliacea (L.) Vahl 283 93.4
Echinochloa crus-galli (L.) P. Beauv. 236 77.9
Paspalum conjugatum Berg. 89 29.4
Cynodon dactylon (L.) Pers. 48 15.8
Echinochloa colona (L.) Link 43 14.2
Monochoria vaginalis (Burm. f.) Presl. 22 7.3
Cyperus iria L.
18 5.9
Panicum repens L.
4 1.3
Echinochloa oryzoides (Ard.) Fritsch 3 1.0
Sporobolus africanus (Poir.) Rob. & Tourn. 3 1.0
lschaemum rugosum Salisb. 1 0.3
Miscellaneous 32 10.6
Weed management practices of rice farmers in Iloilo Philippines 149

major weeds reported by Iloilo farmers in an earlier farmer survey (Estorninos and
Moody 1982).
Among these weeds, farmers found E. crus-galli to be the most destructive. Other
weeds considered destructive were E. colona (L.) Link. Cynodon dactylon (L.) Pers.,
F. miliacea, and Cyperus iria L.
Origin of rice weeds
Some 63.4% of the farmers interviewed believed that rice weeds were soil-borne
rather than seed-borne. This perception contrasts with that of Nueva Ecija farmers,
who generally thought that weed seeds from different Echinochloa species were seedborne
(Fujisaka 1993). Other farmers thought that weed seeds were carried by irrigation
water (30%), mixed with rice seeds (27.1%), came with fertilizer (8.6%), were
carried by animals (2.6%), were blown in by the wind (1.7%), were brought by insects
(1.9%), and were mixed with the haystack (0.3%).
Weed management
Although farmers mentioned a variety of control measures. herbicide application was
the dominant approach, practiced by 94.1% of the farmers. followed by hand weeding
(76.9%) and water management (55.8%). Only a few farmers (0.6%) mentioned
plowing.
Herbicide use. In aggregate terms, farmers indicated the timing of their herbicide
application. Almost all of the farmers reported that they applied herbicide at the seedling
stage, followed by about 25% who treated their crops at the vegetative stage.
Table 3 shows that farmers reported making a total of 422 herbicide treatments. Among
the herbicides reported, pretilachlor was the most commonly used at the seedling
stage (39.3%), followed by piperophos + 2,4-D (31.2%). At the vegetative stage,
piperophos +2,4-D was the main herbicide applied (28.9%), followed by 2,4-D (26.3%)
(Table 3). It should be noted that this herbicide use pattern was reported by farmers in
a survey. To verify this, we need to probe using techniques such as focus group interviews
to solicit further information from farmers.
For herbicide application frequencies, only 5.9% of the farmers did not apply
any herbicide. The most common frequency was only one treatment in the 1995 dry
season (51.2%), whereas 36.6% of the farmers made two, 4.6% made three, and 1 %
made four treatments.
Water management. Some 55.8% of the farmers applied water management to
control weeds. Of these, 52.7% described the method they used as applying herbicide
on dry soil, then introducing water a few days later or merely flooding the field (40.2%).
The interval between herbicide application and flooding varied greatly among farmers.
Some farmers (9.5%) first flooded the field, then drained it, and then applied
herbicides and fertilizers. Among those who used water management, 92.3% carried
it out at 0–20 d after seeding (DAS). Only a few (7.6%) used water management to
control weeds beyond this stage. In the respondents’ households, water management
was mainly done by hired labor (61.7%) and the farmers themselves (38.6%). Other
150 Moody et al

Table 3. Herbicides applied by rice farmers at two different crop stages, Iloilo, dry season
1995.
Herbicide applications at each stage
Herbicide Seedling Vegetative Total
No. % a No. % No. %
Bensulfuron-methyl 1 0.3 0 0.0 1 0.2
Butachlor 41 11.8 4 5.3 45 8.1
Butachlor + propanil 26 7.5 3 4.0 29
7.2
MCPA
8 2.3 13 17.1 21 6.2
Oxadiazon 13 3.8 0 0.0 13 5.5
Piperophos + 2,4-D 108 31.2 22 28.9 130 28.7
Pretilachlor 136 39.3 12 15.8 148 34.7
Thiobencarb + 2,4-D 2 0.6 0 0.0 2 0.5
2,4-D 8 2.3 20 26.3 28 6.5
No answer 3 0.9 2 2.6 5 1.7
Total 346 100.0 76 100.0 422 99.3
Total % 81.9 18.1 100.0
a Percentages for column.
farmers said that the spouse (18.8%) and children (18.2%) were responsible for water
management.
Hand weeding. Many farmers (48.5%) performed their weeding operation at 21–
40 DAS, followed by (37.8%) who did so at 0–20 DAS. A few (5.6%) indicated that
they also weeded their ricefield >41 DAS. Labor allocation for weeding followed the
pattern for water management, with hired labor (78.5%) and the farmers themselves
(48.9%) doing most of the work. Spouses (25.3%) and children (24.9%) also pitched
in with weeding tasks on the respondents’ farms.
Attitudes toward rice seeds
As rice seeds are relevant to weed management, some indications of farmers’ attitudes
toward rice seeds were obtained by using an attitude scale, where farmers were
asked to indicate their degree of agreement or disagreement with a set of attitude
statements concerning rice seeds.
Seeds from private seed growers do not require additional cleaning. By indicating
disagreement with this statement, 60.4% of the farmers seemed to be aware that
seeds from private seed growers would still require additional cleaning. Only 34.9%
believed otherwise. A few (4.6%) expressed no opinion on this attitude statement.
Injection and contaminants in rice seeds do not decrease yield. Evidently, a
large proportion of the farmers (70%) believed that infection and contaminants could
decrease yield. Only 25.7% agreed with this attitude statement. A few (4.3%) had no
opinion.
Seed cleaning is a waste of time. More than half of the respondents (53.1%)
disagreed with this statement, which alludes to the value of seed cleaning. On the
other hand, 46.2% of the farmers believed that seed cleaning is a waste of time.
Weed management practices of rice farmers in Iloilo, Philippines 151

Winnowing removes all infected seeds. As a technique to remove infected seeds,
winnowing was considered to have some benefit by only 46.9% of the respondents.
Some 47.2% disagreed and 5.9% had no opinion.
Neighbors and friends should exchange seeds among themselves. Most of the
farmers interviewed (93.1 %) believed in seed exchange among neighbors and friends.
Only a few either disagreed (4.3%) or had no opinion (2.6%).
Discussion
Intensification in rice production and changes in planting method have resulted in
changes in weed populations, the emergence of new weed problems (including weedy
forms of rice), and increased herbicide use.
Farmer innovations
Most traditional methods of weed control were developed through trial-and-error,
natural selection, and keen observation. Maurya (1989) notes that even in the absence
of on-farm research involving outsiders, farmers regularly innovate and select their
own appropriate technologies. When farmers are seen as experimenters and innovators,
other views also change. More dynamic and flexible research processes building
on farmer-research interactions and supporting farmers’ innovations become possible
(Rhoades 1989).
Experimenting by farmers has long been insufficiently recognized. Farmers are
keen to seek solutions to old and new problems through experimentation (Rhoades
and Bebbington 1991). Farmers experiment to satisfy curiosity, solve problems, adopt
technology (Rhoades 1989), and improve production practices (Litsinger 1993). Crop
production technology of modem rice farmers is an assimilation of traditional practices
into modem farming (Goodell 1984). Iloilo farmers have modified recommended
crop production practices and altered input use (herbicide rates) and application timing.
Farmers applied a second or third herbicide if the previously applied herbicide
failed to control all the weeds. Applying piperophos + 2,4-D and pretilachlor at the
vegetative stage is not a recommended practice.
It has been increasingly realized that farmers have, through long experience,
evolved many useful practices. It is clear, then, that we should be ready to learn from
farmers (Sanghi 1987, cited in Gupta 1989). Merely documenting farmer practices is
not enough. We have to identify the scientific basis of these practices and link it with
their rationality (Gupta 1989).
Herbicides
Fajardo and Moody (1990) noted that farmers innovate in their approach to chemical
weed control in wet-seeded rice by using herbicide combinations, sequential herbicide
applications, nonconventional application methods, varying application times,
and different water management strategies. Effectiveness was the most important factor
in the selection of a herbicide, with price being the next most important consideration
(Estorninos et al 1995).
152 Moody et al

The pressing need to raise yields and maintain profits on a progressively limited
land base has paved the way for herbicide use in Asian rice production. Farmers are
left with little choice but to reduce labor and production costs, particularly for the
most labor-intensive tasks, such as planting and weeding. As a result, herbicides are
being substituted widely for manual labor as a method of weed control. This trend has
been reinforced by the spread of direct-seeded rice, which requires chemical weed
control in the early stages of crop growth to prevent substantial losses, and by a decreasing
ability in some systems to manage weeds through water control as a consequence
of diminishing water supplies and deteriorating irrigation structures (Naylor
1994). Labor unavailability, changes in crop establishment methods, and economics
are the major forces that drive farmers to become herbicide-dependent.
Farmers in most countries in South and Southeast Asia apply herbicides at less
than the recommended rate (Navarez and Moody 1979, Abeyratne et al 1984, van de
Fliert and Matteson 1990). This may be an economic measure, but most claim that
effective weed control is achieved (Navarez and Moody 1979). In addition to credit
constraints, such a behavior could also be caused by risk aversion (Horowitz and
Lichtenberg 1994). On the other hand, chemical companies may have calculated a
safety margin for farmers who use lower amounts of herbicide and for those whose
land preparation and water control are less than desirable (Mabbayad and Moody
1992).
In a number of experiments conducted at the International Rice Research Institute
and in farmers’ fields in the Philippines, rates of application of preemergence
herbicides, such as butachlor and pretilachlor + fenclorim, could be reduced by up to
50% of those recommended without a loss in effectiveness or a reduction in crop
yield (Mabbayad and Moody 1985, Castin et a1 1992, Pablico and Moody 1992).
Farmers in Iloilo used herbicides alone or in combination with hand weeding for
weed control. Hand weeding is done after herbicide application to control weeds not
killed by the herbicides, reduce competition, reduce weed seed reserves in the soil,
and increase yield. With increases in the real wage rate, the price of herbicides has
fallen over time, providing an incentive to use herbicides for weed control in wetseeded
rice (Pandey 1995). Hand weeding is at least five times more expensive for
weed control in wet-seeded rice (Moody 1994b). Farmers apply a herbicide again,
and sometimes make a third application, to obtain better weed control and to eliminate
the need for hand weeding (Heong et al 1995). Guyer and Benjamin (1983)
reported that piperophos + 2,4-D gave good weed control at any time between transplanting
and 4 wk after transplanting.
In South and Southeast Asia, in the next decade, weed control methods in rice
will not differ greatly from those of today. But the area treated with herbicides will
increase because of expanded irrigated area planted to direct-seeded rice and an increase
in herbicide use in less favorable environments. In addition, more effective
herbicides will be used and lower rates will be applied (a shift to herbicides that
require lower application rates for the same effect is already taking place), resulting
in a reduction in the total amount of active ingredients applied, although the number
of applications per field per year is unlikely to change. More emphasis will be placed
Weed management practices of rice farmers in Iloilo, Philippines 153

on less persistent, postemergence herbicides (to allow farmers to wait and see the
nature and extent of developing weed problems) (Moody 1994a). Farmers may, however,
opt for prophylactic (preemergence) applications if they believe that
postemergence control is too risky (Pandey and Medd 1990).
As Williams (1992) succinctly put it, “An axiom that would apply to the use
situation might be ‘as little as possible and as much as necessary.”’ This will require
clear information on the components of the weed flora, the effects of certain levels of
infestation on crop performance, and the effectiveness and cost of potential means of
control. The goal will be to reduce rates and frequencies of herbicide applications in
combination with other practices to achieve the degree of weed control needed. It is
possible to reduce the recommended rate because it is often based on worst-case situations
for the most-difficult-to-control weed species. If a farmer knows the weed species
in his fields and the required rate for controlling each species. then the recommended
rate can often be lowered without a significant loss of effectiveness.
Seed banks
The presence of most weed seeds in the soil can be explained by a combination of
their seed production capacity, longevity characteristics, and species abundance in
the standing community (Regnier 1995). In an agricultural context, weed flora is almost
synonymous with seed banks and, often, the greatest proportion of the plant
community is buried as a seed bank (Harper 1977).
On the basis of a seeding rate of 100 kg ha -1 , rice seeds containing 10 weed seeds
per kilogram would result in 1,000 weed seed species per hectare. In contrast, Kim
and Mercado (1987) reported that the average number of weed seeds in the top 15 cm
of lowland ricefields cropped twice per year was 497,050 m -2 and that in fields cropped
three times per year it was 298,520 m -2 . Thus, annual weed populations are maintained
in arable fields primarily by their viable seeds in the soil (seed bank).
Agricultural land usually carries a large reservoir of weed seeds in the soil. This
number varies from 600 to 496,000 m -2 , which represents up to 85% of the weed
population (Jensen 1969). But only a few species contribute to the bulk of the seed
bank (Kropac 1966).
If there are 100 plants m -2 of weeds and each plant produces 10,000 seeds, at
harvest there will be 1 million seeds m -2 . To maintain the infestation rate of 100 plants
m -2 , a survival rate of 0.01 % is needed. Even for plants with a low seed production, a
low survival rate maintains the infestation (Koch 1974).
If new seeds could be prevented from reaching land with 10,000 seeds m -2 and
50% of these seeds were lost annually to germination and other causes, after 5 yr
there would still be 313 seeds m -2 . Interestingly, if there were 100,000 seeds m -2 , it
would take only 8 yr to reduce the number of seeds to 391 m -2 . If 10% of these germinated
and grew to maturity, and each produced 325 seeds, this would bring back the
number of weed seeds in the soil to the original figure of 10,000 m -2 in one season.
154 Moody et al

Seed health
Removal of weed seeds before maturation to reduce weed problems in the following
crop (Fajardo and Moody 1995) or to reduce contamination of rice seed (Rao and
Moody 1990) is a common practice. To try to ensure seed purity, farmers commonly
rogue off-types before harvest. This will alleviate but not totally eliminate the problem.
A further reduction in off-types would be expected if they were removed before
pollen shedding. Ho (1985) reported that a farmer who planted 1.4 ha was able to
reduce Echinochloa crus-galli weed density from more than 100 panicles m -2 to fewer
than 15 panicles m -2 in the subsequent season by spending about 60 h removing the
panicles throughout the season.
Weeds that mature and shed seed before the crop tend to spread more slowly and
are more localized than weeds that mature at the same time as the rice crop. Early
maturing weeds are spread mostly by the movement of equipment and in water and
air. Weeds that mature with the crop tend to spread as a contaminant in the harvested
crop and in subsequently planted seed (Bayer 1991). The main route for weeds between
farmers is probably through contaminated crop seed.
Seed contaminated with weed seeds may introduce new species to a given field
or add to an existing weed population. If there are 500 weed seeds kg -1 of rice seed
and the seeding rate is 100 kg ha -1 , 50,000 weed seeds will be planted per hectare.
This is equivalent to 5 weed seeds m -2 . Echinochloa crus-galli, which is a powerful
competitor with rice, requires control at densities of 5–10 plants m -2 (Smith 1988).
Weed species in upland ricefields in transmigration areas in South Kalimantan
are the same as those in rural areas of Java, Sumatra, and Kalimantan, Indonesia.
implying that the weed seeds may have been brought along in the planting materials
carried by the transmigrants (Wirjahardja et al 1979).
Contaminated seed is thought to be responsible for the introduction of a number
of exotic rice weeds into France. Among those proliferating are Cyperus difformis, C.
eragrostis, Lindernia dubia, Ammania coccinea, Heteranthera limosa, and H.
reniformis. Echinochloa spp. and red rice also cause problems (Jauzein 1991).
The use of impure rice seed is responsible for the rapid spread of Ischaemum
rugosum in Surinam (Dirven and Poerink 1955) and Echinochloa crus-galli and E.
colona in Malaysia (MARDI 1983).
Red rice or rice with a red-colored pericarp comes from an off-type that is considered
a weed by rice producers and processors. It usually enters a field through
impure seed and then becomes a continual problem through volunteer reinfestation
(USDA 1973).
The advantages of using good-quality seed include higher yield, maintenance of
seed purity (elimination of cultivar mixtures), and higher germination percentage,
resulting in better (more uniform) stand establishment and fewer weed and disease
problems. Farmers in Central Luzon, Philippines, also stated that seedlings from healthy
seeds competed better with weeds and were more tolerant of adverse environments
(Diaz et al 1995). Certified rice seed users in northern Mindanao, Philippines, obtained
25% more yield than farmers who used seed from the previous harvest (Ramos
1987).
Weed management practices of rice farmers in Iloilo, Philippines 155

Farmers in the Philippines use a number of methods to clean rice seed before
planting (Fajardo and Moody 1995), with the efficiency of the method varying from
farmer to farmer, by method and by contaminant. Winnowing, which farmers believe
to be the most efficient method for removing contaminants (Diaz et al 1995), is the
most widely used cleaning method. Fajardo and Moody (1995), however, reported
that it removed only 58% of Echinochloa glabrescens seeds from contaminated rice
seed. In contrast, flotation, another popular seed-cleaning method, removed 87% of
the E. glabrescens seeds. A combination of winnowing + flotation was superior to
either method alone, removing 95% of the weed seeds.
By using combinations of seed-cleaning methods, farmers can remove most weed
seed contaminants in rice seed. This will prevent, to a large extent, the perpetuation of
weeds in fields and the introduction of new weed species.
The most effective means of dealing with weed seed contaminants in crop seeds
for planting are prevention and decontamination. Rao and Moody (1995) determined
the effect of soaking rice seed contaminated with weed seeds in low concentrations of
herbicide solutions on weed control and rice growth. They found that soaking rice
seed contaminated with I. rugosum in 10 ppm of butachlor + safener for 24 h resulted
in a 98% kill of I. rugosum seedlings without a significant reduction in rice seedling
survival. They estimated the cost of soaking 100 kg of rice seed in 300 L of 10 ppm
butachlor + safener solution to be US$0.08. Peudpaichit et al (1985) and Mabbayad
and Moody (1992) reported that seed treatment with herbicides (bensulfuron-methyl
and pretilachlor + safener) appeared to be a promising application technique for controlling
weeds in wet-seeded rice. But possible injurious effects on nontarget organisms
such as the applicator, birds, and soil-inhibiting mammals need to be examined
carefully.
References
Abeyratne F, Janssen J, Naylor P, O’Reilly C, Parfait G, Rader C, Satorre L, Suwandi. 1984.
Farming system in the H3 old settlement area of the Mahaweli Development Project, Sri
Lanka. ICRA Bull. 17. Wageningen, Netherlands. 51 p.
Adhikarya R. 1994. Strategic extension campaign: a participatory-oriented method of agricultural
extension. Rome (Italy): FAO.
Akobundu IO. 1987. Weed science in the tropics: principles and practices. Chichester: John
Wiley & Sons.
Alstrom S. 1990. Fundamentals of weed management in hot-climate peasant agriculture. Crop
Production Science II. Uppsala (Sweden): Swedish University of Agricultural Sciences.
Bayer DE. 1991. Weed management. In: Luh BS, editor. Rice production. Vol 1. 2nd edition.
New York (USA): Van Nostrand Reinhold. p 287-309.
Caseley JC. 1994. Herbicides. In: Labrada R, Caseley JC, Parker C, editors. Weed management
for developing countries. FAO Plant Production and Protection Paper 120. Rome: FAO. p
83-123.
Castin EM, Pablico PP, Moody K. 1992. Effect of water overflowing from a herbicide-treated
rice field on herbicide performance. Trop. Pest Manage. 38:30-35.
Day BE. 1972. Nonchemical weed control. In: Pest control strategies for the future. Washington
(D.C., USA): National Academy of Sciences.
156 Moody et al

Delouche JC. 1988. Weeds as rice seed contaminants. In: Rice seed health. Los Baños (Philippines):
International Rice Research Institute. p 179-187.
Denning GL, Jayasuriya SK, Huey BA. 1983. Constraints to the adoption of new weed control
technology in rice. In: Proceedings of the Conference on Weed Control in Rice, 31 Aug-4
Sep 1981, IRRI and the International Weed Science Society. Los Baños (Philippines):
IRRI. p 345-360.
Diaz CP, Hossain M, Luis JS, Paris TR. 1995 Knowledge, attitude and practice of seed management
technologies in rice farming in Central Luzon. Philipp. J. Crop Sci. 19(2):87-99.
Dirven JGP, Poerink HJ. 1955. Weeds in rice and their control in Suriname. Trop. Agric. 32:115-
123.
Doll JD. 1994. Dynamics and complexity of weed competition. In: Weed management for
developing countries. FAO Plant Production and Protection Paper 120. Rome: FAO. p 29-
34.
Edward JC, Srivastava RN, Singh RB. 1963. Weeds of the Allahabad Agricultural Institute
Campus. Allahabad Farmer 37(2):1-44.
Estorninos LE Jr, Janiya JD, Moody K. 1995. Survey of herbicide usage of wet seeded rice
farmers. Paper presented at the 16th Annual Conference of the Pest Management Council
of the Philippines, 2-6 May 1995, Baguio City, Philippines.
Estorninos LE Jr, Moody K. 1982. Farmers’ weed control practices in rainfed wetland rice
(Oryza sativa) growing areas in Iloilo, Philippines. Philipp. J. Weed Sci. 9: 18-28.
Fajardo FF, Moody K. 1990. Weed control and related cultural practices for wet-seeded rice
(Oryza sativa L.) in Guimba, Nueva Ecija. Philipp. J. Weed Sci. 17:51-64.
Fajardo FF, Moody K. 1995. Farmers’ rice seed cleaning methods in Nueva Ecija, Philippines
and their effectiveness. In: Proceedings of the 15th Asian-Pacific Weed Science Society
Conference, Tsukuba, Japan.
Fujisaka S, Guino RA, Lubigan RT, Moody K. 1993. Farmers’ rice seed management practices
and resulting weed seed contamination in the Philippines. Seed Sci. Technol. 21:149-157.
Goodell G. 1984. Challenges to international pest management research and extension in the
Third World: do we really want IPM to work? Bull. Entomol. Soc. Am. 30(3):18-26.
Gupta AK. 1989. Scientists’ views of farmers‘ practices in India: barriers to effective interaction.
In: Chambers R, Pacey A, Thrupp LA, editors. Farmer first. London (UK): Intermediate
Technology Publications.
Guyer R, Benjamin B. 1983. Influence of water management and spacing on weed cover and
herbicidal activity of Rilof-H in transplanted rice. In: Proceedings of the 9th Asian-Pacific
Weed Science Society Conference, Manila, Philippines. p 369-375.
Harper JL. 1977. Population biology of plants. London (UK): Academic Press.
Heong KL, Teng PS, Moody K. 1995. Managing rice pests with less chemicals. GeoJournal
35(3):337-349.
Ho NK. 1985. Weed problems in the direct seeded and volunteer seedling fields in the Muda
area. In: MADA Monog. NO. 42. Alor Setar (Kedah, Malaysia): Muda Agricultural Development
Authority. p 16-25.
Ho NK, Zuki I, Othman AB. 1990. The implementation of strategic extension campaign on
integrated weed management in the Muda area. In: Proceedings of the 3rd International
Conference on Plant Protection in the Tropics. 6:234-241.
Horowitz JK, Lichtenberg E. 1994. Risk-reducing and risk-increasing effects of pesticides. J.
Agric. Econ. 45:82-89.
Jauzein P. 1991. Eclipta prostata (L.): a weed of rice fields in the Camargue. Monde des Plantes.
86(440):15-16.
Weed management practices of rice farmers in Iloilo, Philippines 157

Jensen HA. 1969. Content of buried seeds in arable soil in Denmark and its relation to true
weed population. Dansk Botanisk Arkiv. 27:1-56.
Kim JS, Mercado BL. 1987. Magnitude and distribution of weed seeds in lowland rice soil.
Philipp. Agric. 7050-59.
Koch W. 1974. Developments in herbology. Paper WP/31. Presented at the Conference on
Plant Protection in Tropical and Sub-Tropical Areas, 4-15 Nov 1974. Manila (Philippines):
Bureau of Plant Industry.
Kropac Z. 1966. Estimation of weed seeds in arable soil. Pedobiologia 6:105-128.
Labrada R, Caseley JC, Parker C. 1994. Weed control in the context of integrated pest management.
In: Weed management for developing countries. FAO Plant Production and Protection
Paper 120. Rome: FAO. p 1-8.
Litsinger JA. 1993. A farming systems approach to insect pest management for upland and
lowland rice farmers in tropical Asia. In: Altieri A, editor. Crop protection strategies for
subsistence farmers. Boulder (Colorado, USA): Westview Press. p 45-101.
Mabbayad MO, Moody K. 1992. Herbicide seed treatment for weed control in wet-seeded rice.
Trop. Pest Manage. 38:9-12.
Mabbayad MO, Moody K. 1985. Improving butachlor selectivity and weed control in wetseeded
rice. J. Plant Protect. Trop. 2:117-124.
MARDI (Malaysian Agricultural Research and Development Institute). 1983. Annual report
for 1982. Bumbong Lima, Kepala Batas, Seberang Perai, Malaysia.
Maurya DM. 1989. The innovative approach of Indian farmers. In: Chambers R, Pacey A,
Thrupp LA, editors. Farmer first. London (England): Intermediate Technology Publications.
Moody K. 1994a. Weed management in rice. In: Labrada R, Caseley JC, Parker C, editors.
Weed management for developing countries. FAO Plant Production and Protection Paper
120. Rome: FAO. p 249-256.
Moody K. 1993. Weed control in wet-seeded rice in tropical Asia. Extension Bulletin No. 364.
Taiwan: ASPAC Food & Fertilizer Technology Center.
Moody K. 1991. Weed management in rice. In: Pimentel D, editor. Handbook of pest management
in agriculture. 2nd edition. Boca Raton (Florida, USA): CRC Press. p 301-328.
Moody K. 1994b. Wet-seeded rice in the Philippines: farmer practices. Proceedings of an International
Symposium in Omagari, July 1995, Tohoku National Agricultural Experiment
Station, Omagari, Japan.
Mortimer AM. 1994. The classification and ecology of weeds. In: Labrada R, Caseley JC,
Parker C, editors. Weed management for developing countries. FAO Plant Production and
Protection Paper 120. Rome: FAO. p 11-26.
Navarez DC, Moody K. 1979. Farmers’ weed control practices in rainfed lowland rice growing
areas in Manaoag, Pangasinan, Philippines. Philipp. J. Weed Sci. 6:55-68.
Naylor T. 1994. Herbicides in Asian rice production: perspectives from economics, ecology,
and the agricultural sciences. Paper presented at a Conference on Herbicide Use in Asian
Rice Production, 28-30 March 1994, Stanford University, California, USA.
Pablico PP, Moody K. 1992. Effect of flooding duration on herbicide performance. Paper presented
at the 23rd Annual Conference of the Pest Management Council of the Philippines,
27-30 April 1992, Tagaytay City, Philippines.
Pandey S. 1995. Socioeconomic research issues on wet seeding. In: Moody K, editor. Constraints,
opportunities, and innovations for wet seeded rice. IRRI Discussion Paper Series
No. 10. Los Baños (Philippines): IRRI. p 73-84.
158 Moody et al

Pandey S, Medd RW. 1990. Integration of seed and plant kill tactics for control of wild oats: an
economic evaluation. Agric. Syst. 34:65-76.
Parker C, Fryer JD. 1975. Weed control problems causing major reductions in world food
supplies. FAO Plant Prot. Bull. 23/24:83.
Peudpaichit S, Ide GS, Poola-or C, Tipyasothi P. 1985. DPX-F5384 herbicide application flexibility
for broadleaf weed control in direct-seeded rice. In: Proceedings of the 10th Asian-
Pacific Weed Science Society Conference, Chiangmai, Thailand. p 114-122.
Ramos LA. 1987. Utilization of certified seeds among rice farmers in northern Mindanao. MS
thesis. University of the Philippines at Los Baños, College, Laguna, Philippines. 160 p.
Rao J. 1968. Studies on the development of tubers in nutgrass and their starch content at different
soil depths. Madras Agric. J. 55:19-23.
Rao AN, Moody K. 1990. Weed seed contamination in rice seed. Seed Sci. Technol. 18: 139-
146.
Rao AN, Moody K. 1995. Seed soaking in herbicide solution for controlling Ischaemum rugosum,
a rice seed contaminant. In: Proceedings of the 15th Asian-Pacific Weed Science Society
Conference, Tsukuba, Japan. p 670-675.
Regnier EE. 1995. Teaching seed bank ecology in an undergraduate laboratory exercise. Weed
Technol. 9:5-16.
Rhoades R. 1989. The role of farmers in the creation of agricultural technology. In: Chambers
R, Pacey A, Thrupp LA, editors. Farmer first. London (UK): Intermediate Technology
Publications.
Rhoades T, Bebbington A. 1991. Farmers as experimenters. In: Haverkort B, van der Kamp J,
Waters-Bayer A, editors. Joining farmers’ experiments. London (UK): Intermediate Technology
Publications.
Soerjani M. 1970. Alang-alang, Imperata cylindrica (L.) Beauv., pattern of growth as related to
its problem of control. BIOTROP Bulletin 1. Bogor, Indonesia: Regional Center for Tropical
Biology.
Smith RJ Jr. 1988. Weed thresholds in southern U.S. rice. Oryza sativa. Weed Technol. 2:232-
241.
SAS (Statistical Analysis System). 1985. SAS user's guide: statistics. Version 5 Edition. Cary
(N.C., USA): SAS Institute, Inc.
USDA (United States Department of Agriculture). 1973. Rice in the United States: varieties
and production. Agric. Handb. 289, rev. ed. Washington (D.C., USA): Agricultural Research
Service.
van de Fliert E, Matteson PC. 1990. Rice integrated pest control training needs identified through
a farmer survey in Sri Lanka. J. Plant Protect. Trop. 7:15-26.
Watanabe H, Azmi Man. 1995. Weed seed bank, longevity and dormancy in relation to weed
management in rice. In: Rapusas HR, Heong KL, editors. Workshop report: weed management
in rice production. p 253-278.
Williams RJ. 1992. Management of weeds in the year 2000. In: Kadir AASA, Barlow HS,
editors. Pest management and the environment in 2000. Wallingford (UK): CAB International.
Wirjahardja S, Dekker RJ, Utomo IH, Eussen JHH, Laumonier EK, Megia R. 1979. The biology
of important weeds (distribution, taxonomy, ecology and physiology) in rice fields.
Biotrop Newsl. 30:7-8.
Weed management practices of rice farmers in Iloilo. Philippines 159

Notes
Authors’ addresses: K. Moody, Agronomy, Plant Physiology, and Agroecology Division, International
Rice Research Institute (IRRI), Los Baños, Philippines; M.M. Escalada, Department
of Development Communication, Visayas State College of Agriculture, Baybay, Leyte,
Philippines; KL Heong, Entomology and Plant Pathology Division, IRRI, Los Baños, Philippines.
Acknowledgments: The authors would like to thank the Swiss Agency for Development and
Cooperation (SDC), through the Rice IPM Network based at IRRI, for funding the farmer
survey reported in this paper, Dr. T.W. Mew and Dr. M. Hossain for sharing their expertise,
and our research collaborators for making the survey possible.
Citation: Heong KL, Escalada MM, editors. 1997. Pest management of rice farmers in Asia.
Manila (Philippines): International Rice Research Institute.
160 Moody et al

CHAPTER 11
Pest management perceptions
and practices of farmers growing rice
and vegetables in Nueva Ecija,
Philippines
K.L. Heong, A.A. Lazaro, and G.W. Norton
Pest management perceptions and practices of farmers growing both
rice and vegetables were studied. Results show that farmers’ practices,
particularly early season and unnecessary applications of insecticides,
were similar in both rice and vegetables. Farmers sprayed
more frequently and used more toxic insecticides in vegetables, however,
than in rice. The main targets of insecticide sprays in both
crops, particularly during the early crop stages, were insect pests
causing leaf damage, and highly toxic compounds were used to control
these pests. Farmers’ practices usually reflected their percep
tions of pests and their damage.
Introduction
Farmers’ knowledge, attitudes, and practices in managing rice pests have been well
documented. Some of this information is presented in the other chapters in this book.
But a similar documentation of farmers’ management of vegetable pests is minimal.
A general observation is that, in vegetables, farmers tend to be more pesticide-dependent.
This heavy pesticide use is due partly to the high “cosmetic” value demanded by
consumers, forcing farmers to deliver damage-free produce. This market pressure to
produce high-quality farm products encourages vegetable growers to adopt measures
to keep their harvest free from visible insect damage.
In developing countries, vegetable farmers usually have poor technical support.
Consequently, widespread pesticide misuse in vegetable farming has been reported
(Adalla 1990, Adalla and Hoque 1991, Bernardo 1992, Medina 1987, Vos et al 1993,
Tjornhom et al 1996). In rice growing, an analysis showed that more than 80% of
insecticide sprays that farmers applied in a season could be considered as misuse
(Heong et al 1994b). The main reasons for such misuse were farmers’ misperceptions
of pests, their overestimation of potential damages and losses (Heong et al 1994a,
Lazaro et al 1993), and attitudes favoring insecticide use (Lazaro and Heong 1995).
This knowledge gap, which has made farmers victims of insecticide misuse, is well
161

described by Bentley (1989) for Honduran farmers and may well be true for fanners
handling pest problems in general.
Pest management research in the past 30 years has focused on technology generation,
rather than on solving farmers’ problems. The approach has been described
as “technology push.” To solve farmers’ problems would require a “farmers’ needs
pull” approach (Conway and Barbier 1990). Here, the need to better understand how
farmers perceive problems, their beliefs, and factors influencing their decisions and
practices is essential. This paper compares the pest management knowledge, perceptions,
and practices of farmers who were growing both rice and vegetables in San
Jose, Nueva Ecija, Philippines. The paper compares the types of pesticides used by
the farmers, their respective pest targets, and pest management decisions in the two
crops.
San Jose is a municipality located in the province of Nueva Ecija, the rice bowl of
the Philippines. As a principal trading and commercial center of the province, it has a
total land area of 18,725 ha, with a population of 92,083 (52% males). More than
50% of the total land area is agricultural (9,628 ha), with 6,644 ha (69%) irrigated and
2,982 ha (31%) rainfed. In 1990, San Jose had a farmer population of 4,800 in 38
villages called “barangays.” About 62% of the farmers in San Jose were managing
farms of 1–3 ha. More than two-thirds (68%) of the farmlands were rented or leased
and only 22% were fully owned.
Rice is the main crop in San Jose. About 3,380 farmers, or 71%, cultivate rice in
irrigated land, whereas the rest produce rice in the rainfed environment. Other field
crops grown are maize, root crops, legumes (mungbean, peanut, cowpea, and pole
string beans), vegetables (cabbage, pechay, mustard, lettuce, bittergourd, eggplant,
squash, tomato, patola), and spices (onion, garlic, sweet pepper, and hot pepper).
Most farmers in San Jose grow two crops per year—rice during the wet season
and vegetables during the dry season. Rice is usually planted in June, July, or August,
and vegetables in October, November, or December. Sometimes, two vegetable crops
are grown in the dry season, with the second crop starting in January, February, or
March. Intercropping is quite common, particularly for those farmers who grow more
than one vegetable crop per season. For instance, rice is often compared with eggplant,
which has a 7-mo duration. Farmers’ main constraint to agricultural production
is the irregular supply of water for irrigation, which explains why most farmers grow
vegetables during the dry season, as vegetables require less water than rice.
Methods
Data on farmers’ knowledge, attitudes, and practices in managing pests attacking rice
and vegetables were gathered through a survey in the barangays Tayabo, Santo Tomas,
Abar lst, Kita-Kita, Palestina, and Sibut in June 1994. The farmers were selected
randomly from the list of names obtained from the local agricultural office in each
barangay. The survey questionnaire was translated into Tagalog, the local language
spoken by the farmers, and pretested to determine whether the respondents could
162 Heong et al

understand it. After pretesting, the final questionnaire was modified accordingly. Farmer
responses were coded and entered in a spreadsheet using Microsoft Excel ® .
In addition to the survey, focus group interviews were also conducted in small
farmer groups, where the authors were able to probe for additional answers. A total of
300 farmers were interviewed, but, for this paper, only the data from 230 farmers who
grew both rice and vegetable crops were used. Cross-tabulations of various variables
were generated using PC SAS to compare farmers’ pest management perceptions and
practices in rice and vegetables. The general linear models procedure in SAS (SAS
1985) was used for the analyses of variance. For tests of association between variables,
we used the Pearson chi-square in the frequency procedure in SAS.
Results
Background information
Most of the farmers in the survey area planted modern high-yielding rice varieties
during the 1993 wet season, followed by vegetables for market during the dry season
of 1993-94. More than 90% of the farmers grew onions. Other vegetable crops grown
were tomato, pepper, eggplant, string beans, garlic, bittergourd, and pechay. Onion
(Allium cepa Lind.) was the main vegetable crop of farmers who grew more than one
vegetable type per year. This was usually grown for the local market and for farmers’
own consumption. Locally known as sibuyas from the Spanish cebolla, it is also the
most important bulb crop and one of the most profitable vegetables in the Philippines.
The farmers interviewed usually managed rice farms of 2 ha or less and grew
vegetables on small plots. During the dry season, farmers used small portions of their
rice farms for vegetables, leaving the rest fallow. Land preparation consisted of alternate
plowing and harrowing two to three times after harvesting a rice crop. In another
method, without plowing the rice straw, weeds are cut close to the ground. For onion,
farmers usually start seedlings on raised beds and later transplant them to the field.
Pesticide use
A large proportion of the farmers used pesticides in their rice (95%) and vegetable
(97%) production during the respective growing seasons. The total number of pesticide
applications in one season was higher in vegetables, with an average of 5.3 (SD
= 3.82) sprays per farmer per season compared with an average of 3.3 (SD = 1.71)
sprays for rice. Nearly all pesticide applications were insecticides (Table 1).
Spray frequency distributions in the two crops were also different (Fig. 1). Most
farmers used 3 applications in each crop, but 26% applied more than 6 sprays in
vegetables compared with only 3% of the farmers who applied more than 6 in rice.
Farmers’ spray frequencies in rice were significantly related to their spray frequencies
in vegetables (Pearson c = 21.25, probability = 0.012), implying that farmers
who used more sprays in rice tended to use more sprays in vegetables as well.
Many farmers interviewed were using methyl parathion, monocrotophos, endosulfan,
and cypermethrin in both rice and vegetables (Table 2). About 47% of the farmers
used methyl parathion, a WHO category Ia insecticide (Table 3), on vegetables, whereas
Pest management practices of farmers in Nueva Ecija, Philippines 163

Table 1. Comparison of pesticide applications of farmers growing
rice and vegetables in San Jose, Nueva Ecija, Philippines,
June 1994.
Pesticide applications Rice Vegetables
Total number of sprays 762 1,201
in one season
Farmers who sprayed 94.8 97.4
last season (%)
Mean number of sprays per 3.3 5.3
farmer per season
Insecticides (%) 99.5 98.3
Fungicides (%) 0.4 1.5
Herbicides (%) 0.1 0.2
Fig. 1. Distribution of farmers’s spray frequencies in rice and vegetables, San Jose.
28% used it on rice. More farmers used monocrotophos (53%) and endosulfan (50%)
on rice than on vegetables (33% and 31%). Fewer farmers used carbamates, except
for methomyl, which was used by 10% and 8% of the farmers in rice and vegetables,
respectively. Methomyl is another highly toxic chemical (classified as WHO category
Ib). More farmers used pyrethroid insecticides in rice than in vegetables. The most
popular chemical was cypermethrin, used by 53% of the farmers in rice and 46% in
vegetables.
164 Heong et al

Table 2. Insecticides used by farmers a in rice and vegetables, San Jose, Philippines.
Insecticides
Rice
WHO
category b No. %
Vegetables
No. %
Organochlorines
Endosulfan II 108 49.5 69 30.8
Endrin la 1 0.5 0 0.0
DDT II 0 0.0 1 0.5
Organophosphates
Monocrotophos Ib 116 53.2 75 33.5
Diazinon II 1 0.5 1 0.5
Azinphos-ethyl Ib 7 3.2 7
3.1
Methamidophos Ib 1 0.5 1
0.5
Malathion III 4 1.8 15 6.7
Methyl parathion la 61 28.0 106 47.3
Profenofos II 1 0.5 2 0.9
Chlorpyrifos II 8 3.7 4 1.8
Chlorpyrifos + fenobucarb - 10 4.6 7 3.1
Azinphos + ethyl + fenobucarb - 1 0.5 0 0.0
Carbamates
Carbofuran Ib 10 4.6 6 2.7
Carbaryl II 1 0.5 12
5.4
Methomyl
Ib 21 9.6 19 8.5
lsoprocarb II 2 0.9 0
0.0
Fenobucarb II 2 0.9 0
0.0
Pyrethroids
Deltamethrin II 32 14.7 53
Cypermethrin II 115 52.8 103
Lambda cyhalothrin
-
21 9.6 10
Fenvalerete II 0 0.0 1
23.7
46.0
4.5
0.5
Others
Ethofenprox - 2 0.9 1
0.5
Buprofezin V 1 0.5 0 0.0
Farmers who sprayed (no.) 2 18 224
a Multiple responses. Farmers may have used more than one Insecticide.
b la = extremely hazardous; Ib = highly
hazardous; II = moderately hazardous; III = slightly hazardous; V = unlikely to present acute hazard in normal
use; - = unclassified.
Source: CIRAD 1991.
Table 3. Classification of insecticides used by farmers a in rice and vegetables,
San Jose, Philippines.
WHO classification
by health hazard
Rice
Vegetables
No. % No. %
la: Extremely hazardous 62 28.4 106 47.3
Ib: Highly hazardous 123 56.4 84 37.5
II: Moderately hazardous 182 83.5 160
71.4
III: Slightly hazardous 4 1.8 15
6.7
V: Unlikely to present acute hazard 1 0.5 0 0.0
in normal use
Unclassified 34 15.6 18 8.0
Farmers who sprayed (no.) 218 224
a Multiple responses. Farmers may have used more than one type of insecticide.
Source: ClRAD 1991.
Pest management practices of farmers in Nueva Ecija, Philippines 165

The correlation of the ranks of percentage data in Table 2 was highly significant
(r = 0.789, P = .0001) using Spearman’s rank-order correlation, indicating a strong
linear relationship. This indicates that the proportion of farmers using each insecticide
was more or less the same in both rice and vegetables.
Spray targets of farmers
The most common spray targets in rice were “worms” (33% of sprays), followed by
stem borers (20%), hoppers (15%), and rice bugs (11%) (Table 4). Worms, or uod in
Tagalog, was the generic word used to describe lepidopterous species, such as the
leaffolders Cnaphalocrocis medinalis, Nymphula depunctalis, Spodoptera spp.,
Nuranga aenescens, and Pelopidas mathias. These were also the primary targets of
farmers’ applications during the early crop stages (seedling and tillering). At the flowering,
milking, and soft dough stages, rice bugs were the main targets. The main
Table 4. Target pests of rice and vegetables in San Jose, Philippines.
Target pests Sprays on rice Sprays on vegetables
and diseases
No. % No. %
Worms 284 33.3 369 47.8
Stem borers 173 20.3 4 0.5
Hoppers 125 14.7 12 1.5
(mostly planthoppers)
Rice bugs 91 10.7 7 0.9
Moths 81 9.5 10 1.3
Golden apple snail 38 4.4 1 0.1
Thrips 15 1.7 331 42.9
Tungro 10 1.2 3 0.4
Mosquito 7 0.8 4 0.5
Fungi 4 0.5 2 0.3
Ladybird beetle 4 0.5 2 0.3
Mole cricket 3 0.3 2 0.3
Whorl maggot 2 0.2 1 0.1
Ants 2 0.2 11 1.4
Grasshopper/locust 5 0.6 -
Caseworm 4 0.5 -
Leaffolder 2 0.2 -
Hopperburn 1 0.1 -
Rats 1 0.1 -
Bullfrog 1 0.1 -
Dumping off - 5 0.6
Bulb rot - 3 0.4
Aphids - 2 0.3
Weeds -
1 0.1
166 Heong et al

chemical compounds used by farmers for rice bugs were monocrotophos, cypermethrin,
and endosulfan.
For vegetables, worms (48%) and thrips (43%) were the main pest targets of
insecticide sprays (Table 4). These sprays were used throughout the vegetable growing
season. Worms in onions were generally Spodoptera litura, whereas other vegetables
had a variety of lepidopterous species. At the leafy stage, thrips were more
commonly sprayed. The most common compounds used to control these pests were
cypermethrin, methyl parathion, and endosulfan.
For both crops, worms had been farmers’ main insecticide targets, although the
pest species were different. In rice, the main species farmers referred to as worms was
the rice leaffolder, C. medinalis, whereas in vegetables the species was S. litura.
Insecticide use perceptions
Nearly all the farmers interviewed believed that it was necessary to use chemicals to
control pests in both rice (97%) and vegetables (98%). Many farmers (67% in rice
and 52% in vegetables) believed that early season sprays within the first 4 wk after
crop establishment were necessary. In some cases, control decisions (32% in rice and
27% in vegetables) depended on damage signs. Most farmers (96% in rice and 98% in
vegetables) also believed that insecticide use would increase crop yields and disagreed
(86% in rice and 89% in vegetables) that insecticides could induce the development
of other pests.
The crop yields of farmers who sprayed within the first month after crop establishment
were compared with yields of those who sprayed late (after the first month).
In addition, yields of farmers who sprayed frequently (more than 3 times) were compared
with those who did not for both rice and vegetables separately. Yields in all
cases for both vegetables and rice were not significantly different (early spraying in
rice: F = 0.88, P = 0.52; frequent spraying in rice: F = 0.72, P = 0.71; early spraying
in vegetables: F = 0.94, P = 0.48; frequent spraying in vegetables: F = 0.88. P = 0.59).
This implies that farmers who sprayed early did not necessarily obtain higher yields
than those who sprayed late (after the first month), and those farmers who sprayed
more than 3 times did not obtain higher yields than those who sprayed less, in vegetables
as well as in rice production.
Discussion
Farmers’ pest management perceptions and practices for both rice and vegetables in
San Jose appeared to be similar. Insect pests seem to be their primary concern as
illustrated by the high proportions of insecticide sprays used in a season. Farmers
differ, however, in the number of sprays used. For rice, farmers usually used 3 sprays,
whereas they used 5 for vegetables. This may reflect differences in crop values as
well as degrees of pest infestations. The types of compounds used were also similar.
The main chemicals used were cypermethrin, methyl parathion, monocrotophos, and
endosulfan. The last three chemicals had been banned and placed under restricted use
by the Fertilizer and Pesticide Authority (FPA) of the Philippines in June 1994, but
Pest management practices of farmers in Nueva Ecija, Philippines 167

they were still used by most farmers in rice and vegetables. This situation was also
similar to that found in Leyte in 1991 (Heong et al 1994a).
In rice, worms, stem borers, and planthoppers were the common pests targeted by
farmers, whereas in vegetables worms and thrips were the main pests. Because damage
caused by these insects tends to be highly visible to farmers, this might have
stimulated decisions to spray. Uod was the generic name farmers used for most leaffeeding
lepidopterous larvae, and these were the prominent targets of insecticide sprays
during the seedling and vegetative stages in both rice and vegetables. In rice, these
worms were mainly leaffolders, cutworms, caseworms, and stem borers, whereas in
vegetables they were the onion maggot, Delia antiqua, and cutworm, S. litura. Farmers
often attributed leaf damage, whether visible or not, to worms. Some of these
damage symptoms may well be caused by onion nematodes.
A common vegetable pest farmers target for their sprays is thrips (Thrips tabaci
Lindeman), a species found on many crops such as onion, tomato, eggplant, and cucumber.
Adults and nymphs cause damage by sucking the plant sap from younger
leaves in the growing plants, causing silvery shoots. When damage is severe, the entire
field may become silver-colored. Thrips tend to be most damaging when they feed
during the early bulbing stage of the onion crop. High populations of thrips have been
reported to reduce both yield and quality of onions and farmers tended to believe that
insecticide treatments were always necessary. But there is insufficient research quantifying
pest-yield relationships, which can be used to help develop a control strategy.
Onion crops often recover from thrips injury without the aid of insecticides. Farmers
in San Jose mainly used methyl parathion and cypermethrin to control thrips.
Pest damages most visible to farmers were those made on the leaves of the crop.
Farmers would often attribute low yields to such damage symptoms. In rice, plants
can compensate for foliar damage and yields are unaffected (Miyashita 1985, Rubia et
al 1996). Because farmers’ control decisions in most cases are based on perceived
rather than real losses, there is a high tendency to overestimate losses (Lazaro et al
1993). When these perceptions prevail, many of the spray decisions farmers make
may not result in any economic gain. Farmers’ responses to pests in both rice and
vegetables are clearly due to extreme risk aversion. Part of the problem comes from
the huge knowledge gap existing between what farmers need to know and what they
presently know about pest management. This knowledge gap appears to be even greater
for vegetable production. This could account for similarities between insecticide use
patterns by farmers in the two crops.
Farmers perceived that insecticides are necessary to maintain a healthy crop and
to obtain high yields. Our analyses, however, showed no relationship between timing
of sprays and yield, or between frequency of applications and yield. Because spray
applications had not significantly increased farmers’ crop yields, economic returns
from their spraying seem questionable. Using a logit analysis to assess socioeconomic
factors that influence pesticide misuse on vegetables in San Jose, Tjornhom et al (1997)
found that increased misuse is associated with visits by chemical company representatives
or by Department of Agriculture technicians.
168 Heong et al

Improving farmers’ knowledge of pests and natural enemies of rice and vegetable
crops through training may help farmers make better management decisions. if
the knowledge is applied. Another approach is to aim at changing farmers’ perceptions
of pests, their damages, and insecticide use. The prevailing perceptions are that
highly visible pests are damaging and that leaf damage would lead to significant yield
losses, that insecticides are essential inputs to obtain high yields, that insecticides
need to be applied at the early crop stages, that the highly toxic chemicals are more
effective in controlling pests, and that early and more frequent insecticide applications
would lead to higher yields. In rice, farmer experiments successfully changed
farmers’ perceptions and practices regarding rice leaffolders (Heong and Escalada
1997). Such farmer participatory approaches should be explored to enhance decision
making in vegetable growing as well.
References
Adalla CB. 1990. Integrated pest management, extension and women project. Terminal Report
Volume. Department of Entomology, College of Agriculture, University of the Philippines
Los Baños, College, Laguna. 217 p.
Adalla CB, Hoque MM. 1991. Integrated nutrient and pest management, extension and women
project. B-1. Evaluation and monitoring of post phase I farmer cooperators. Year I annual
report (1990-1991), University of the Philippines Los Baños and International Development
Research Centre. p 18-25.
Bentley JW. 1989. What farmers don’t know can’t help them: the strengths and weaknesses of
indigenous technical knowledge in Honduras. Agric. Human Values 6:25-31.
Bernardo EN. 1992. Vegetable insect pest management: status, constraint, controversy and
prospect. Philipp. Agric. 75(3&4):1-19.
Conway GR, Barbier EB. 1990. After the green revolution: sustainable agriculture for development.
London: Earthscan Publications Ltd.
Heong KL, Escalada MM. 1997. Perception change in rice pest management: a case study of
farmers’ evaluation of conflict information. J. Appl. Commun. 81(2):3-17.
Heong KL, Escalada MM, Lazaro AA. 1994b. Misuse of pesticides among rice farmers in
Leyte, Philippines. In: Pingali P, Roger P. editors. Impact of farmer health and the rice
environment. Boston (Mass., USA): Kluwer Press.
Heong KL, Escalada MM, Vo Mai. 1994a. An analysis of insecticide use in rice: case studies in
the Philippines and Vietnam. Int. J. Pest Manage. 40(2):173-178.
CIRAD (International Cooperation Centre in Agronomic Research for Development). 1991.
Regional agro-pesticide index. Vol. 1. Asia. 3rd ed. Bangkok (Thailand): Franco-Pacific.
Lazaro AA, Heong KL. 1995. Analysis of factors influencing farmers’ insecticide sprays. Int.
Rice Res. Notes 20(2):20.
Lazaro AA, Rubia EG, Almazan LP, Heong KL. 1993. Farmers’ estimates of percent whiteheads
(WH). Int. Rice Res. Notes 18(4):31.
Medina CP. 1987. Pest control practices and pesticide perceptions of vegetable farmers in Loo
Valley, Benguet, Philippines. In: Tait J, Napompeth B, editors. Management of pests and
pesticides: farmers’ perceptions and practices. Boulder (Colo., USA): Westview Press. p
150-157.
Pest management practices of farmers in Nueva Ecija, Philippines 169

Miyashita T. 1985. Estimation of the economic injury level in the rice leaf roller, Cnaphalocrocis
medinalis Guenee (Lepidoptera: Pyralidae). I. Relationship between yield loss and injury
of rice leaves at heading or in the grain filling period. Jap. J. Appl. Entomol. Zool. 29:73-
76.
Rubia EG, Heong KL, Zalucki M, Gonzales B, Norton GA. 1996. Mechanisms of compensation
of rice plants to yellow stem borer, Scirpophaga incertulas (Walker) injury. Crop
Protect. 15:335-340.
SAS (Statistical Analysis System). 1985. SAS user’s guide statistics. Version 5 edition. Cary
(N.C., USA): SAS Institute, Inc.
Tjornhom J, Norton GW, Heong, KL, Talekar NS, Gapud V. 1997. Socioeconomic characteristics
and pesticide misuse in Philippine onion production. Philip. Entomol. 11:139-149.
Vos JGM, Nurtika N, Sumarni N. 1993. An exploratory survey on farmers’ practices and management
of hot pepper (Capsicum spp.) in Java, Indonesia. J. Plant Protect. Trop. 10:91-
109.
Notes
Authors’ addresses: K.L. Heong and A.A. Lazaro, Entomology and Plant Pathology Division,
International Rice Research Institute, P.O. Box 933, 1099 Manila, Philippines. G.W. Norton,
Department of Agricultural and Applied Economics, Virginia Polytechnic Institute and
State University, Blacksburg, VA 24061, USA.
Acknowledgments: The authors wish to thank the Integrated Pest Management Collaborative
Research Support Program (IPM CRSP, USAID Grant # LAG-G-00-93-0053-00) for supporting
the baseline survey, which was coordinated by the Philippine Rice Research Institute
(PhilRice) in Muñoz, Nueva Ecija, and the International Rice Research Institute (IRRI)
in Los Baños, Philippines. In particular, we are grateful to Dr. Santiago R. Obien, director
general of PhilRice, and Dr. Vic Gapud, IPM CRSP project coordinator, for providing the
personnel to assist in the survey. We are also very grateful to Mr. Bernard Canapi for his
important contribution to this research, the students who helped us interview the farmers,
and the farmers who gave us their time.
Citation: Heong KL, Escalada MM, editors. 1997. Pest management of rice farmers in Asia.
Manila (Philippines): International Rice Research Institute.
170 Heong et al

CHAPTER 12
Pest management practices
of rice farmers in Sri Lanka
L. Nugaliyadde, T. Hidaka, and M.P. Dhanapala
Nearly 330 rice farmers in seven administrative districts of Sri Lanka
were interviewed to determine their pest management practices and
knowledge of insect pests and natural enemies. Most of the farmers
(95%) were unaware of the pest-resistant characteristics of commonly
grown rice varieties.
During the main season (October to March) of 1995-96, 79% of
the farmers reported infestations of the following insect pests:
leaffolder (22%), rice bug (17%), brown planthopper (16%), stem
borer (14%), thrips (5%), cutworm (5%), gall midge (4%), and armyworm
(4%). Most farmers (86%) applied insecticides to control these
pests, 7% did nothing, and the rest used traditional methods. Farmers
applied insecticides an average of 1.2 times per season, of which
83% were to control pests instead of preventing infestations. Some
42% of the insecticide applications were made at the seedling stage
and 40% at the vegetative stage. The most commonly used insecticides
were carbofuran (10%), chlorpyrifos (7%), phenthoate (7%),
and monocrotophos (5%).
Most of the farmers (92%) believed that leaf-feeding insects
would cause damage to rice and 61% thought that spraying should
be done early to avoid severe crop losses. Although 82% of the farmers
knew the natural enemies of rice pests, only 17% knew their role
in the rice ecosystem.
Introduction
Rice in Sri Lanka
As the staple food in Sri Lanka, rice is the most important food crop in the country. It
is also the livelihood of more than 1.8 million farm families, and more than 30% of
the total labor force is directly or indirectly involved in the rice sector. From 1991 to
1995, the average annual rough rice production was about 2.24 million t. At the present
consumption rate for rice (100 kg capita -1 yr -1 ), Sri Lanka would require about 3.5
million t of rice to feed its population of about 20 million in the year 2000 (DOASL
171

1995a). Because Sri Lanka is reaching a stage at which further expansion in rice area
will not be possible, the alternative is to give more emphasis to increasing the production
potential per unit area of land and finding ways to achieve this on-farm through
appropriate technology (DOASL 1995b).
The main problems now facing the rice sector are low and stagnating yield (3.2 t
ha -1 ), escalating production costs ($175 t -1 rough rice), diminishing profitability ($12.50
month -1 irrigated ha -1 ), poor grain quality, and the high market price of polished rice
($0.30–0.46 kg -1 ) (DOASL 1995a). The increasing prices of inorganic fertilizers, increasing
labor costs, and the declining labor force for rice cultivation had a negative
effect on the number of farmers using improved crop management practices. A need
has therefore arisen to undertake detailed investigations on biological and socioeconomic
constraints to rice production to sustain the gains that have been achieved and
to meet the challenges of declining productivity and profitability in rice (DOASL
1995b). The government therefore developed a comprehensive plan to intensify research
and development efforts to: (1) reduce the cost of production (below $100 t -1 of
rough rice), (2) maintain the quality of the rice environment, (3) increase national
yield from 3.2 to 4.5 t ha -1 , and (4) increase the professionalism of rice farming (DOASL
1995a).
Integrated management of rice pests
The improvement of crop protection technologies that are cheaper, more efficient,
and environmentally sound is becoming urgent. But inadequate crop protection techniques,
the lack of appropriate agronomic practices, insufficient extension services,
and socioeconomic constraints are the main problems associated with implementing
integrated pest management (IPM) techniques (DOASL 1995b). Therefore, emphasis
has been given to developing rice varieties with resistance to the major pests—rice
thrips, gall midge (biotype II), and brown planthopper—and studying the natural control
mechanisms of these pests, analyzing farmers' pest management decision making
and practices, and introducing different IPM approaches to improve farmers' practices
in order to develop sustainable IPM in rice. Furthermore, the research activities
of the Department of Agriculture, Sri Lanka (DOASL), focus mainly on generating
research information on IPM components to help strengthen farmer training
(Kudagamage and Nugaliyadde 1995, Fernando 1996).
Survey sites
Rice ecosystem
Sri Lanka has approximately 730,000 ha of land suitable for rice cultivation. Of this,
41 % is covered by the major irrigation schemes, 25% by the minor irrigation schemes,
and 34% is under rainfed conditions, with a total area sown per year of around 830,000
ha because of double cropping. The average productivity under major, minor, and
rainfed lands is 4.2 t ha -1 , 3.2 t ha -1 , and 2.4 t ha -1 , respectively. Therefore, the irrigated
lands (major and minor) contribute about 75% to the total paddy production
(Pathinayake et al 1991, DOASL 1995a, 1995c).
172 Nugaliyadde et al

Sri Lanka’s ricelands are mostly in the inland valleys and, to a limited extent, on
coastal plains, flood plains, and terraced slopes. Variable seasonal rains and a variety
of soils, elevations, temperatures, and drainage patterns create complex and varying
environments for rice production (Will 1989). The agroclimatic features of the country
are strongly influenced by the interaction of the two monsoons—the northeast
(October-March) and the southwest (May-September)—and the mountainous land
mass in the south-central areas. Climatic zones are grouped on the basis of annual
rainfall into dry (900–1,500 mm), intermediate (1,500–2,200 mm), and wet (>2,200
mm), and on the basis of altitude into low-country (1,000 m) (Fig. 1). A further classification of lands into 24
agroecological regions was made according to altitude, rainfall and its distribution,
major soil groups, elevation, temperature, and vegetation (Panabokke and Kannangara
1975).
The cultivation year is divided into two seasons—main ( maha ) between October
and March and minor (yala) between April and September. The seasons are very
distinct in the dry zone, but become increasingly less distinct toward the intermediate
and wet zones. The northeast monsoon (October to March) is stronger and produces
rains throughout the island. The southwest monsoon (May to September) brings rains
mostly to the wet zone, the southwestern quarter, and the central highlands. As a
result, moisture is sufficient year-round in the wet zone. but dry spells are frequent in
the dry and intermediate zones (Panabokke and Kannangara 1975).
Sampling sites
Seven administrative districts — Anuradhapura, Polonnaruwa, Kurunegala, Matale,
Kandy, Ratnapura, and Hambantota — were selected to represent different
agroecological regions (Fig. 1). Accessibility to the sample areas was also considered
in selecting these districts. Anuradhapura, Polonnaruwa, and Hambantota are mainly
dry zones, whereas the others run across wet, intermediate, and dry zones. These
seven districts cover about 35% of the total ricelands in the country. More than 90%
of the ricelands in Anuradhapura, Polonnaruwa, and Hambantota districts are under
irrigated cultivation compared with those in Kurunegala (63%), Matale (75%), Kandy
(63%), and Ratnapura (63%). A significant proportion of the ricelands in Kurunegala,
Matale, Kandy, and Ratnapura is under rainfed cultivation (Gunatilake and Somasiri
1995).
Although pest problems pertaining to different areas are not documented systematically,
Anuradhapura, Polonnaruwa, Hambantota, and Ampara are endemic regions
for brown planthopper. Rice gall midge infestations are usually high in the dry and
intermediate zones during the main season and in the wet zone in the minor season
(Kudagamage and Nugaliyadde 1995).
Pest management practices of rice farmers in Sri Lanka 173

Fig. 1. Zone and district boundaries of Sri Lanka.
174 Nugaliyadde et al

Method of survey
Selection of farmers
Farmers who cultivate irrigated rice in the dry and intermediate zones and favorable
rainfed rice in the wet zone were selected for the survey. Two or three Agrarian Service
Centers (ASC) were selected on purpose from each administrative district to
represent different agroecological regions. About 25 farmers, selected at random, were
interviewed from each ASC range. A total of 329 farmers from Anuradhapura,
Polonnaruwa, Kurunegala, Kandy, Matale, Ratnapura, and Hambantota were interviewed.
Survey instrument
The survey was designed to describe the sociodemographic profile of rice farmers in
the selected districts, record the agronomic and pest management practices followed
during the main season of 1995-96 (October 1995 to March 1996), and obtain detailed
information about farmers’ knowledge and attitudes on insect pests and natural
enemies of rice and their use of pesticides.
A questionnaire already used by the IPM Network in other Asian countries was
taken as the base document in preparing the survey instrument. The questionnaire
was translated into the Sinhala language and adapted to reflect local situations. The
adapted questionnaire was pretested with 10 farmers from Kurunegala and modifications
were made accordingly.
Survey procedure
To ensure consistency in responses, the enumerators agreed to follow a uniform questioning
and recording procedure. Follow-up discussions helped share experiences and
make necessary adjustments to the survey procedure. The survey was done mainly
during weekends and holidays in order to find as many farmers as possible. Initially,
the interviewers gave farmers a brief introduction to the survey objectives and requested
their cooperation for its success.
Farmers were interviewed within a period of 10 wk (February-April 1996). The
survey began soon after the main season of 1995-96 and continued until the beginning
of the minor season of 1996 to enable farmers to relate observations from previous
seasons and activities more accurately.
Results
Sociodemographic profile of farmers
The ages of selected rice farmers ranged from 60, with a mean and SE of 46.2
± 3.4. About one-fifth of the farmers were below the age of 30, and 4.3% were above
61 (Table 1). Some 58.5% of the farmers were owner-operators and 26.4% were tenants.
Only 9.2% were lessees, 1.1% were hired laborers, and 4.5% fell into other
categories. Of the 329 farmers interviewed, 38.9% had more than 20 yr of experience
in rice farming and 73.2% had cultivated rice for more than 10 yr.
Pest management practices of rice farmers in Sri Lanka 175

Table 1. Sociodemographic profile of rice farmers interviewed
(n=329) (cropping season October 1995-March 1996).
Profile
Age class

86% of the crop was broadcast. Broadcasting was more popular in Hambantota (100%),
Anuradhapura (85%), and Polonnaruwa (80%) than in Matale (39%) and Ratnapura
(25%).
Farmers reported grain yield from 2 to 6 t ha -1 , with a mean and SE of 3.34 &
0.67. About 29% and 28% of the farmers obtained rice yields of 34 and 4–5 t ha -1 ,
respectively. One-third of the farmers obtained yields below 3 t ha -1 , the national
average, and 12% produced more than 5 t ha -1 . More than 70% of the farmers in
Polonnaruwa, Anuradhapura, and Hambantota obtained yields of 4–5 t ha -1 compared
with those in Kurunegala, Ratnapura, Kandy, and Matale of 34 t ha -1 .
Rice farmers' pest management knowledge
Most farmers (95%) were unaware of the pest-resistant characteristics of rice varieties.
Of the 329 farmers interviewed, 79.6% reported at least one pest infestation during
the past season (October 1995-March 1996). The rest (20%) mentioned no pest
damage or ignored low infestation levels. Rice leaffolder infestations were reported
by 22% of the rice farmers. This was followed by rice bug (17%), brown planthopper
(BPH) (16%), stem borer (14%), rats (9%), thrips (5%), cutworm (5%), armyworm
(4%), and gall midge (4%). Fewer than 3% of the farmers reported rice diseases—
blast, bacterial blight, and rice yellowing. About 87% of the farmers (all those who
direct-seeded their crop) reported weeds as their main problem during the past season.
When farmers were asked to indicate the most important pest problem, 33%
mentioned leaffolder, 14% rice bug, 13% rats, 10% BPH, and 7% stem borer. The
second most important pest problems mentioned were rice bug (31%), stem borer
(25%), BPH (16%), rats (5%), thrips (4%), cutworm (4%), and gall midge (4%).
Of the 262 farmers who reported insect pest problems, 86% have applied insecticides
to control them, 7% did nothing, 3.4% used baiting, 2% used water management,
and a small portion resorted to hand picking and other traditional methods,
including spraying neem seed extracts.
Pesticide application pattern
The farmers who used chemical control made a total of 402 pesticide (insecticides
and fungicides) applications, or an average of 1.2 applications per farmer. Of these
applications, 83% were made to control pests and 15% to prevent pest infestations;
the remaining 2% were made with uncertainty. Furthermore, most of the pesticide
applications were made at the seedling stage (42%) and vegetative stage (40%). Only
18% of the applications were made at the reproductive stage of the crop.
During the main crop season, farmers spent from

Table 2. Main target pests of pesticides used by farmers and number of applications.
Chemicals LF a BPH PB SB TH GM AW/ Rats N %
CW
Endosulfan b 6 - 4 - - - - - 10 2.5
Chlorpyrifos 21 1 8 - - - - - 30 7.4
Diazinon
- 6 4 - - 3 - - 13 3.2
Dimethoate
3 - 1 - 3 - - - 7 1.7
Fenthion 6 - - - - - - - 6 1.5
Phenthoate
14 - - - 3 - 11 - 28 6.9
Methamidophos b 13 - - - 2 - 1 - 16 1.5
Malathion b 1 2 - - 1 - - - 4 0.9
Monocrotophos b 7 5 - 1 - - 6 - 19 4.7
Benfuracarb - - - - 2 3 - - 5 1.2
Carbofuran 3%G - - 29 - 1 12 - - 42 10.4
Fenobucarb - - - - - 6 - - 6 1.5
Cyfluthrin
2 - - - - - - - 2 0.4
Brodifacoum - - - - - - - 3 3 0.7
Tebufenozide 1 - - - - - - - 1 0.2
Sulfur b 1 - - - - - - - 1 0.2
Unknown 34 68 37 37 15 - - 2 193 47.9
Total 109 82 83 38 27 24 185 386
a LF = leaffolder, BPH = brown planthopper, PB = paddy bug, SB = stem borer, TH =thrips, GM = gall midge, AW/
CW = armyworm and cutworm. b lnsecticides not recommended for rice.
Farmers' selection of insecticides for rice pest control
The pesticides (insecticides and fungicides) used by farmers during the main crop
season fall into different groups (Table 2). Of these, endosulfan, methamidophos,
monocrotophos, malathion, cyfluthurin, fungicides, and sulfur are not recommended
for rice pest control (endosulfan and monocrotophos have been banned in Sri Lanka
since 1995). Some products are used exclusively in public health care (cyfluthrin and
malathion), but some farmers (n=12) used fungicides for insect control. More interestingly,
48% of the pesticides used were classified as unknown because the farmers
were unable to remember the name of the pesticide.
The most commonly used insecticides were carbofuran (10.4%), chlorpyrifos
(7.4%), phenthoate (6.9%), monocrotophos (4.7%), methamidophos (1.5%), and
diazinon (3.2%). Furthermore, farmers often do not follow the recommended application
rates (DOASL 1995d). During the survey, we observed that more than 75% of
the farmers never followed the recommended dosage and dilution. Farmers gave many
explanations for not following the recommendations indicated on the label of the
pesticide container. In addition, the higher cost of pesticides and labor caused farmers
to use lower dosages at the first sign of insect infestation.
Farmers’ perceptions about leaf-feeding insects
Almost all farmers mentioned the leaffolder as one of the most damaging leaf-feeding
insects of rice. Farmers recognized grasshoppers, cutworms, armyworms, and
caseworms as leaf-feeding insects. Farmers in Polonnaruwa were more aware of leaffeeding
insects than those in other districts and were able to identify at least three
178 Nugaliyadde et al

Table 3. Farmers’ perceptions about leaf-feeding insects (LFI) of rice (main season October
1995 March 1996).
Criterion
District a (farmers responding, %)
Ap Pol Kur Kay Mat Rat Ham Av
Average number of LFI 2.1 3.1 2.0 1.2 1.5 1.3 2.3 1.9
reported by farmers
LFI cause severe damage 66 69 92 71 69 70 66 71.8
to the crop
Insecticides are required to 60 60 69 77 49 73 63 64.4
control LFI
Spraying for LFI should 72 51 80 48 37 81 72 63.0
be done early
a Ap = Anuradhapura, Pol = Polonnaruwa, Kur = Kurunegala, Kay = Kandy, Mat = Matale, Rat = Ratnapura, Ham
= Hambantota.
Table 4. Farmers’ perceptions about natural enemies of rice pests (main season October 1999
March 1996).
Criterion
District a (farmers responding, %)
Ap Pol Kur Kay Mat Rat Ham Av
Know the presence
of natural enemies 33 68 48 31 50 53 86 52.7
Average number of
natural enemies named
by farmers 2.5 4.5 4.1 3.3 4.6 2.7 5.2 3.8
Know the role of
natural enemies 31 67 40 31 49 52 86 50.8
Know the effect of
insecticides on natural
enemies 30 65 41 26 48 51 85 49.4
Know that killing natural
enemies will increase
pest infestations 10 23 33 10 18 11 15 17.1
a Ap = Anuradhapura, Pol = Polonnaruwa, Kur = Kurunegala, Kay = Kandy, Mat = Matale, Rat = Ratnapura,
Ham = Hambantota.
leaf-feeding insects of rice (Table 3). In all districts. the majority of the farmers (54-
95%) believed that leaf-feeding insects would cause severe damage to rice. Close to
two-thirds of the farmers (64%) across all districts believed that insecticides were
needed to control leaf-feeding insects and that spraying should be done early (63%)
to avoid severe crop losses.
Farmers’ knowledge of natural enemies of rice pests
About 53% of the farmers recognized the presence of a group of organisms called
natural enemies in ricefields (Table 4). In Hambantota, 86% of the farmers knew
about natural enemies. In contrast, farmer awareness of natural enemies was corn-
Pest management practices of rice farmers in Sri Lanka 179

paratively poor in Kandy (31%), Anuradhapura (33%), Kurunegarala (48%), and
Matale (50%).
Almost all of the farmers who recognized the presence of natural enemies could
describe their role in the rice environment. Fifty-one percent of the farmers knew that
natural enemies feed on other insects such as pests. Many farmers were able to name
at least three natural enemies of rice pests (mean 3.8). The most common natural
enemies mentioned by farmers were spiders, dragonflies, birds, snakes, different types
of beetles, and flies.
Fifty percent of the farmers were aware of the adverse effects of insecticides on
natural enemies. Farmer awareness of the ill effects of insecticides was high in
Hambantota (85%) and Polonnaruwa (65%). Farmers’ responses to further questions
on the effect of insecticide applications on pest populations were poor. This was evident
in the low proportion of farmers who answered a question on pest resurgence
after insecticide application. Only 17% of the farmers agreed that killing natural enemies
will increase pest infestations.
Discussion
Prior to the beginning of the FAO Intercountry Rice IPM Program in the early 1980s,
insect pest control for rice was mainly based on pesticides. Emphasis was also given
to developing and introducing varieties resistant to major insect pests and understanding
the natural control mechanisms of the insect pest complex of rice (Wickremasinghe
1980, Fernando 1996). In addition to the extensive promotional activities of the pesticide
industry, the well-established extension service network of the DOASL helped
disseminate pesticide recommendations to farmers more effectively. This could be a
major factor contributing to farmers’ overreliance on insecticides for rice pest control.
During the survey, some farmers were observed to consider pesticides (insecticides)
as essential components of rice cultivation; farmers purchased them at the beginning
of the season along with other inputs such as seed, fertilizer, etc. Farmers
would go through financial hardships just to buy pesticides, which are not subsidized
in Sri Lanka. Furthermore, some farmers were found to mix granular insecticides
(usually one-fifth of the recommended dose) with fertilizers at the first topdressing.
Therefore, to implement an ecologically based pest management program, it is essential
to correct farmers’ perceptions that pesticides are important and an essential component
of rice farming.
The improvements made so far on pesticide recommendations have been encouraging.
The DOASL has constantly updated the recommendations. In 1971, the DOASL
covered two organochlorines (DDT and BHC) and six organophosphates (diazinon,
fenthion, malathion, monocrotophos, phenthoate, and trichlorfon) (DOASL 1971).
This recommendation was revised in 1979 to add four carbamates (fenobucarb, carbaryl,
carbofuran, and propoxur) (DOASL 1979). A 1983 revision deleted all organochlorines
from the pesticide recommendation (DOASL 1983). In 1995, the recommendation
was further revised to delete all WHO class Ia and class Ib products and to
include safer new products (DOASL 1995d).
180 Nugaliyadde et al

The IPM program in rice, launched by the DOASL in the early 1980s, first emphasized
improving farmers’ pest management practices by upgrading their knowledge
on pests, natural enemies, and pesticide application patterns. Economic threshold
values for each pest were introduced to justify pesticide applications (Kudagamage
and Nugaliyadde 1995). But the acceptability of pest management based on threshold
levels has received much criticism (Matteson et al 1984).
Because IPM is recognized as a national policy in Sri Lanka, emphasis has been
given to improving farmers’ pest management decision making through farmer field
schools (DOASL 1995a, Fernando 1996). The results of this program seem encouraging.
Farmers in areas where IPM programs have been active over the past decade
seem to be well aware of IPM approaches.
As indicated by van de Fliert (1987), we also observed a critical lack of farmer
knowledge on recommended crop management practices. Awareness of varieties is
important for planning and executing appropriate crop management practices, including
fertilizer application, to realize yield potential. Furthermore, farmers should possess a
sound knowledge of insect pests, including their natural control agents and influence
on yield. among other aspects, to effectively make management decisions. The many
knowledge gaps we observed among farmers may have prevented them from making
effective management decisions.
This was evident because of the prominence farmers gave to insect pests affecting
their crops. Similar to the observations of Escalada et al (1992), Heong et al
(1992, 1994), and Vo Mai et al (1993), we also found that farmers seem to equate
visibility of a pest with its importance. As in other Asian countries, farmers in Sri
Lanka consider leaffolder and paddy bug as the most damaging pests to the rice crop.
Farmers‘ perceptions of leaffolder damage and the need for immediate control have
been reported to be widespread in many Asian countries (Heong et al 1994). On the
other hand, the specific odor and distinct appearance in the field and illuminated places
in the house make paddy bug a prominent pest for farmers. Almost all farmers believe
that the paddy bug causes extensive yield losses to their crops, which stimulates them
to use chemical control.
Although the survey did not study farmers’ fertilizer application patterns, we
observed many variations in fertilizer use. Many farmers seem less interested in supplying
the essential nutritional requirements to seedlings (to incorporate organic matter
or to supply basal fertilizer), which are vital for early growth and to help withstand
pest infestations. Farmers considered nutritional disorders, commonly observed in
low-fertile soils, as pest problems requiring insecticide applications. Therefore, a high
percentage of farmers applied pesticides at the early vegetative stage of the crop.
We also noted that farmers’ knowledge of correct pesticide use is poor, including
selection of the correct pesticide, dosage, dilution, timing, and application techniques
(van der Fliert 1987). This lack of knowledge on proper pesticide use could lead to
environmental as well as socioeconomic problems for rice farmers. Therefore, efforts
should be made to wean farmers from the misuse of pesticides.
Because of the activities of the IPM program, farmer awareness of natural enemies
was found to be high in Hambantota and Polonnaruwa. Efforts should there-
Pest management practices of rice farmers in Sri Lanka 181

fore be made to introduce similar programs in other areas as well. Nevertheless, farmers’
inability to understand pest resurgence because of pesticides; the existence of an
ecological balance among pests, natural enemies, and neutral fauna in ricefields; and
the relationship between pest densities and yield should be taken into account when
developing strategies to improve farmers’ pest management decisions.
References
DOASL (Department of Agriculture Sri Lanka). 1971. Chemical pest control recommendation
for paddy. Peradeniya (Sri Lanka): Agriculture Extension Division, DOASL. 13 p.
DOASL (Department of Agriculture Sri Lanka). 1979. Major crop pests and their control with
pesticides. Peradeniya (Sri Lanka): DOASL. 46 p.
DOASL (Department of Agriculture Sri Lanka). 1983. Major crop pests and their control with
pesticides. Peradeniya (Sri Lanka): Extension and Communication Center, DOASL. 51 p.
DOASL (Department of Agriculture Sri Lanka). 1995a. Work plan for the development of the
rice sector in Sri Lanka. Peradeniya (Sri Lanka): DOASL. 19 p.
DOASL (Department of Agriculture Sri Lanka). 1995b. Five year work plan for the Rice Research
and Development Institute. Peradeniya (Sri Lanka): DOASL. 11 p.
DOASL (Department of Agriculture Sri Lanka). 1995c. Cost of cultivation of agricultural crops—
maha 1993/94. Peradeniya (Sri Lanka): Socio-economic and Planning Center, DOASL.
55 p.
DOASL (Department of Agriculture Sri Lanka). 1995d. Major crop pests and their control with
pesticides. Peradeniya (Sri Lanka): Extension and Communication Center, DOASL. (Unpublished.)
Escalada MM, Lazaro AA, Heong KL. 1992. Early spraying by rice farmers in Leyte, Philippines.
Int. Rice Res. Newsl. 17:27-28.
Fernando MHJP. 1996. Sri Lanka National IPM Programme: a country brief. Proceedings of
the programme advisory committee meeting, FAO Intercountry Programme for IPM in
Asia, 6-9 Feb 1996. Hyderabad (India): FAO. 10 p.
Gunatilake GA, Somasiri S. 1995. Rice growing ecosystems. In: Amarasiri SL, Nagarajah S,
Perera BMK, editors. Rice Congress 1990. Proceedings of the Rice Symposium, 3-4 Sept
1990. Kandy (Sri Lanka): DOASL. p 39-54.
Heong KL, Escalada MM, Lazaro AA. 1992. Pest management practices of rice farmers in
Leyte, Philippines. Los Baños (Philippines): International Rice Research Institute. 57 p.
Heong KL, Escalada MM, Vo Mai. 1994. An analysis of insecticide use in rice: a case study in
the Philippines and Vietnam. Int. J. Pest Manage. 40(2):173-178.
Kudagamage C, Nugaliyadde L. 1995. Present status and future direction of insect pest management
in rice. In: Amarasiri SL, Nagarajah S, Perera BMK, editors. Rice Congress
1990. Proceedings of the Rice Symposium, 3-4 Sept 1990. Kandy (Sri Lanka): DOASL.
p 39-54.
Matteson PC, Altieri MA, Gagne WC. 1984. Modification of small farmer practices for better
pest management. Annu. Rev. Entomol. 29:383-402.
Panabokke CR, Kannangara RPK. 1975. The identification and demarcation of the agro-ecological
regions of Sri Lanka. Proceedings Section B. Ann. Session Assoc. Advmt. Sci. Sri
Lanka Assoc. Advmt. Sci. Colombo 31(3):49.
182 Nugaliyadde et al

Pathinayake BD, Nugaliyadde L, Sandanayake CA. 1991. Direct seeding practices for rice in
Sri Lanka. In: Direct seeded flooded rice in the tropics. Manila (Philippines): IRRI. p 77-
90.
van de Fliert E. 1987. Knowledge, attitudes, practices and information source relating to pest
control by rice farmers in Sri Lanka. Thesis submitted to the Agriculture University,
Wageningen, Netherlands. 229 p.
Vo Mai, Thu Cuc NT, Hung NQ et al. 1993. Farmers’ perception of rice pest problems and
management tactics used in Vietnam. Int. Rice Res. Newsl. 18:31.
Wickremasinghe N. 1980. Fifteen years of progress in rice pest control. In: Rice Symposium
1980. Peradeniya (Sri Lanka): DOASL. p 49-76.
Will H. 1989. Rice environmental classification in Sri Lanka. UNDP/FAO Rainfed Rice Research
and Development Project, SRL/84/024. Central Rice Breeding Station, Batalagoda,
Ibbagamuwa, Sri Lanka. 40 p.
Notes
Authors’ addresses: L. Nugaliyadde, entomologist, Rice Research and Development Institute
(RRDI); T. Hidaka, Japan International Cooperation Agency IPM expert assigned to RRDI;
M.P. Dhanapala, director of RRDI, Department of Agriculture, Batalagoda. Ibbagamuwa,
Sri Lanka.
Acknowledgments: This study was undertaken as a collaborative research between the International
Rice Research Institute and the Department of Agriculture, Sri Lanka, and we are
grateful to Dr. K.L. Heong, IRRI entomologist, for the help extended. Mr. R.M. Hearth
Band, economist, and Mr. M.M.P. Muthunayake, economic assistant, Rice Research and
Development Institute, helped prepare the survey instrument and guided in planning and
conducting the survey. Mr. Ajith Wijesinghe, Mr. Lal Gunawardene, Mr. Udaya Kumara
Ekanayake, Mr. S. Shamel Weniwella, Ms. Inn Samanmale. Ms. Puspa Damayanthi Kumari.
and Ms. Tharanga Hearth helped conduct the survey. Data analysis was done by Ms. Puspa
Damayanthi Kumari and Mr. Jayantha Senanayake.
Citation: Heong KL, Escalada MM, editors. 1997. Pest management of rice farmers in Asia.
Manila (Philippines): International Rice Research Institute.
Pest management practices of rice farmers in Sri Lanka 183

CHAPTER 13
Farmers’ perceptions and practices
in weed management in Thailand
L. Meenakanit and P. Vongsaroj
A survey of 300 rice farmers was conducted to determine their perceptions
of and practices in weed management. The survey involved
eight provinces of the Chao Phraya basin in the central plain and two
provinces of the Pak Panang basin in the south of Thailand. The data
were collected through personal interviews using a pretested questionnaire.
The most important weeds present in the areas surveyed were
grassy weeds—namely, Echinochloa crus-galli (L.) P. Beauv. and
Leptochloa chinensis (L.) Nees, and broadleaf weeds such as
Monochoria vaginalis (Burm. f.) Presl. Chemical application was the
main weed control method farmers used and 2,4-D and butachlor +
propanil were the herbicides used. Most of the farmers knew that
weeds are the main cause of yield reductions as well as being a
harbor for other pests.
A majority of the farmers expressed favorable attitudes toward
weed management, such as the importance of early detection of
weeds, good land preparation, synchronized planting, and application
of the recommended dosage of herbicides. Among the different
weed management methods, herbicide applications and mechanical
methods were the options known by most farmers.
Introduction
During the past decade, rice growing in the irrigated lowlands in Thailand has gradually
shifted from transplanting to wet seeding. As in other countries, the migration of
farm labor into other sectors such as industry, services, construction, and overseas
labor; lower investment in rice planting; the decline in real prices of rice; and higher
yields of direct-seeded versus transplanted rice have been the major reasons for the
shift in crop establishment method (Kanchanomai et al 1970, Panitchpat 1994). With
the high rate of economic growth and industrialization, farmers face rising farm labor
costs for transplanting and weeding. At the same time, real prices of rice have de-
185

clined. This twin dilemma of rising labor costs and falling rice prices has been cited
as an important explanation for the expansion of direct seeding in Asian countries
(Jirstrom 1996, Denning et al 1983).
But direct seeding has adverse effects: increased weed problems, a shift in the
dominant species to grassy weeds, higher pest and disease incidence because of high
planting density to prevent weeds, and difficulties in carrying out recommended cultural
practices (Moody 1993). The rise in weed infestations in direct seeding results
from the greater competition from weeds, which may grow along with the rice plants.
Besides competing with the rice crop for light, water, and soil nutrients, weeds smother
rice because they are more vigorous (De Datta 1990). The shift in weed flora from
broadleaf weeds and sedges to grassy weeds makes it difficult for farmers to differentiate
weeds from rice because of their age and morphological similarity (Moody 1993).
To control weeds, most farmers rely heavily on herbicides. The availability of
many types and formulations of herbicides in the market has generated problems for
Thai farmers. Besides the higher input cost of herbicides, their misuse and overuse
from improper selection and application are often reported. Knowing the right kind,
rate, and timing of herbicide application as well as the stage of the weeds is important
for effective weed control (Vongsaroj 1991, Choutummatut et al 1994).
The integration of traditional weed control methods and physical methods with
moderate herbicide use has been shown to have advantages over excessive dependence
on herbicides. Misuse of herbicides can lead to shifts in weed populations toward
perennial species and the establishment of weeds that are more difficult to control
or have herbicide resistance (Labrada and Parker 1994, Heong et al 1995a). But
better herbicide performance is achieved when optimum cultural practices, particularly
water management, are used. Day (1972) reported that herbicides should be
applied in conjunction with sound cultural management practices because the chemicals
do not serve as their substitutes. The simultaneous application of a variety of
practices has a synergistic effect on weeds, thus resulting in more effective control
(Heong et al 1995a).
On the other hand, farmers’ knowledge of and attitudes toward other weed control
methods seem to have been overlooked. These nonherbicide control options include
proper land preparation, seed cleaning, water management, and other cultural
practices to reduce weed infestation as well as propagules in the weed seed bank. As
Labrada and Parker (1994) have affirmed, research must shift its focus to include
farmers’ needs in order to encourage more effective adoption of integrated weed management.
This study aimed to identify farmers’ problems in weed control and to understand
their perceptions, attitudes, and practices in weed management in order to design
strategies to help them minimize yield losses from weed infestations. The area
selected was central and south Thailand, the country’s major rice-growing areas.
186 Meenakanit and Vongsaroj

Thailand
Located in Southeast Asia, Thailand is surrounded by Myanmar, Laos, Cambodia,
and Malaysia. This area is under the influence of the southwest monsoon, which lasts
from the middle of May until October. Annual rainfall varies from 1,000 to 2,000
mm, according to place and year. The climate, which is warm and humid, is suited to
agriculture. Geographically, Thailand is divided into four main regions: the north,
northeast, central plain, and the south. The central plain is regarded as the rice bowl of
Thailand, whereas in the north, northeast, and south, rice is cultivated mainly for
home consumption. In 1994, the total rice area was estimated to be 8.9 million ha
(DOAE 1994). The main (wet) season covers 8.6 million ha.
Most of the survey provinces are flat lands situated along river basins and lakes
such as the Chao Phraya River and its tributaries and Bueng Borapet Lake for Nakhon
Sawan; Mae Klong and Pachee rivers for Ratcha Buri: Chao Phaya River for Chainat;
Prachin Bun and Hanuman rivers for Prachin Buri; Chao Praya, Lop Buri, and Pasak
rivers for Ayutthaya; Pasak and Lop Buri rivers for Lop Bun; Pasak River for Saraburi;
Tachin River for Suphan Buri; Tapee, Pak Panang, Pakpoon, and Nakhon Si Thammarat
rivers for Nakhon Si Thammarat; and Songkhla Lake for Phatthalung. Rainfall averages
from 1,042 mm to 2,148 mm, whereas mean temperature ranges from 27.2 °C to
28.8 °C.
Three major rice-growing methods are practiced in Thailand: deepwater, directseeded,
and transplanted. The direct-seeded method is widely practiced in the central,
eastern, and western regions. In wet seeding, pregerminated seeds are broadcast over
a puddled field or in one with standing water. This method reduces labor, but it also
results in heavier weed infestations. To control weeds, farmers use a high sowing rate
of 125–187 kg ha -1 . But this sowing rate produces poor tillering and a thick stand,
resulting in temperature and humidity conducive to spreading disease and increasing
insect pest populations.
Weed problems in rice in Thailand are similar to those of other countries in the
region. The yield reduction from weed infestations ranges from 20% to 80% depending
on weed species, their density, cropping season, and weed control practices. Weed
infestations are usually higher in the wet season than in the dry season. The weeds
generally found in most direct-seeded areas in Thailand are Echinochloa spp.,
Leptochloa chinensis (L.) Nees, Jussiaea linifolia Vahl, Monochoria vaginalis (Burm.
f.) Presl., Sphenoclea zeylanica Gaertn., Cyperus difformis L., and C. pulcherrimus
Willd. ex Kunth, among others (DOA 1989).
Official recommendations for weed management include proper land preparation
and leveling, use of clean seeds, and proper selection and use of herbicides (DOAE
1995). In terms of farmers’ weed control practices, Meenakanit et al (1994) reported
that most farmers plow two times, then harrow and puddle the soil before sowing
seeds. Preemergence herbicides such as pretilachlor, butachlor, or oxidiazon are applied
at 6–10 d after seeding (DAS). Herbicides, usually 2,4-D or its derivatives, are
applied 15–20 DAS to control broadleaf weeds that have survived even after the first
application or moved in from neighboring fields. Chinawong (1993) reported that
Farmers' perceptions and practices in weed management in Thailand 187

farmers still lack sufficient knowledge of herbicide selection and application in Nakhon
Pathom Province and other direct-seeded areas.
Water management after herbicide application is also considered crucial for effective
weed control. Flooding the field within 10 d gives better control than flooding
the field 10 d after sowing (Bhandhufufalk and Hare 1985). Postemergence herbicides
will be more effective if the water level is controlled to submerge the weeds
(Chinawong 1992). A seed rate of 100 kg ha -1 is suitable to minimize weed problems
in direct-seeded rice (Kanchanomai 1981). Time of planting also affects the weed
population. During the rainy season, S. zeylanica and Cyperus difformis occur more
than in the dry season, whereas Echinochloa spp. and L. chinensis occur more in wetseeded
rice.
Methods
Data collection
The personal interview was the main method used to collect data on farmers' perceptions,
attitudes, and practices in rice weed management. The instrument, first written
in English, was translated into Thai and back-translated to English to ensure that the
translation reflected the intended meaning of the terms used.
Questionnaire pretesting
The translated questionnaire was subsequently pretested with rice farmers in Chai
Nat to determine the clarity of the wording and translation of the technical terms
used, the logical sequence of the questions, adequacy of the questionnaire instructions,
and the estimated duration of the interview. Two rounds of pretesting were
undertaken to try out the questionnaire using individual interviews and focus group
interviews as the primary research techniques. In the first pretest, individual interviews
were conducted with rice farmers to solicit their feedback to the prototype
questionnaire. The second pretest involved focus group discussions to further test the
revised instrument. Results of the pretest served as the basis for revisions in the questionnaire
and logistical arrangements for the fieldwork.
The survey was conducted by a team of Department of Agricultural Extension
(DOAE) extension technicians (kaset tambon) and Plant Protection Service Unit
(PPSU) technicians.
Sampling procedure
A total of 300 rice farmers who practiced direct seeding were selected from eight
provinces within the Chao Phraya, Bang Pakong, and Thachin basins in the central
plain. These are Nakhon Sawan, Chainat, Lop Buri, Saraburi, Ayutthaya, Prachin Buri,
Suphan Buri, and Ratcha Buri and two provinces in the Pak Panang basin of Nakhon
Si Thammarat Province and Phatthalung of the Songkhla Lake basin (Fig. 1). In each
province, 30 farmers were selected randomly from two or three villages in one subdistrict.
188 Meenakanit and Vongsaroj

Figure 1. Location map of survey areas in Thailand.
Farmers' perceptions and practices in weed management in Thailand 189

Data processing and analysis
Survey data were encoded using a spreadsheet program and processed in a microcomputer
using the Statistical Package for Social Sciences (SPSS ® ). In most cases,
percentages were based on the total sample. Likewise, where multiple responses were
obtained, the sample size was used to compute the percentages.
Results
Sociodemographic profile of farmers
The ages of rice farmers interviewed ranged from 20 to 80, with more than half belonging
to the 41–60 age bracket. This age distribution reflects the one found in previous
studies in the Philippines (Heong et al 1995b). A large proportion of the farmers
were either owner-operators (60%) or tenants (59%). Only one respondent was a hired
laborer. Most (91%) had 1–4 years of primary schooling. Only 3% reported that they
had not attended school at all and few respondents went to high school (5%) and
college (1%).
Information sources
As in other farm surveys, the extension technician appears to play a key role as a
source of information for rice farmers in Thailand. For farming information, the extension
technician was named as a source by 59% of the farmers, followed by relatives
(13%), other farmers (8%), and PPSU technicians (6%). For advice on pest
management, 73% of the farmers consulted the extension technician. Only 11% reported
that they approached the PPSU technician. Other sources of pest management
advice cited were other farmers (4%), relatives (5%), and the radio (2%). These results
are consistent with findings of earlier studies on farmers' sources of information
on crop protection.
Agronomic practices
Rice varieties planted. Farmers reported planting a number of rice varieties for both
the first and second cropping. For the first crop, Lueng Pratiew, a local variety. was
the cultivar grown by 44% of the respondents. It is taller than most hybrid varieties
and commands a good market value; this explains its popularity among farmers. It
was followed by Chainat (34%) and Supanburi (11%). Other local varieties grown by
farmers are Kao Lueng, Puang Thong Pho Thong, Lueng Yai, Pathum 2, Lueng Tahaeng,
Kao Ta-haeng, Sai Ngern, Opol, Malay, and Kao Samut. For the second crop,
Chainat topped the list of varieties planted by farmers (57%), followcd by Supanburi
(18%), local varieties (15%), and IR23 (6%).
Sources of rice seeds. Farmers’ sources of rice seeds consisted of both personal
and institutional sources. Nearly half of the farmers (49%) obtained their rice seeds
for planting from nearby farmers, usually their neighbors. The popularity of farmers’
seeds has to do with access and cost. Farmers found other farmers’ seeds cheaper and
easier to obtain than government seeds. Almost a third (29%) procured seeds from
190 Meenakanit and Vongsaroj

oth the seed center and DOAE seed exchange project. Some 23% used seeds from
their own stock and 11% got seeds from the Department of Agriculture (DOA). Others
bought seeds from the rice mill (3%), local shop (2%), and cooperative (1%).
Rice area. Most (92.3%) of the rice farmers interviewed had direct-seeded their
crop. For the first crop, 39.4% of the farmers in the sample had rice farms of 3.3–6.4
ha, followed by 39% who managed

Seeding rate. In the first crop, 56% of the farmers used a seeding rate of 63–125
kg ha -1 , followed by 36% who seeded their farms at a rate of 126–187.5 kg ha -1 . Some
6% used a very low rate (187.5 kg
ha -1 ) . The distribution of seeding rates for the second crop closely parallels that of the
first crop, with 46% of the farmers using a rate of 126–187.5 kg ha -1 , followed by 42%
who used 63–125 kg ha -1 of rice seeds. In the dry season. farmers plant a shorter rice
variety. Chainat, which requires a higher seed rate. In contrast, in the wet season,
farmers grow Leung Pratiew, a taller plant type. The data show that, in both seasons,
farmers used a higher seed rate than what has been recommended. perhaps to obtain a
thicker rice canopy that could shade out weeds, to have a better crop stand, and to
compensate for the poor germination of the planting materials.
Cost of farm inputs. Most farmers spent between 400 for it last season.
Cost of pest management inputs. For pest management, farmers seem to have
some difficulty estimating the cost of insecticides and fungicides used as shown by
the large proportion of no answers. For 75% of the farmers, herbicides cost from

Table 3. Types of weeds observed by farmers in
Thailand in 1996.
Weeds
First cropping Second cropping
No. % No. %
Grassy weeds 198 66.0 196 65.3
Broadleaf weeds 8 2.7 8 2.7
Sedges 58 19.3 56 18.7
Ferns 1 0.3 1 0.3
Don’t know 35 11.7 39 13.0
Table 4. Most important weeds observed by farmers in Thailand
in 1996.
Weeds
Farmers
No. %
Grassy weeds 252 73.0
Leptochloa chinensis (L.) Nees 85 33.7
Echinochloa crus-galli (L.) Beauv. 64 25.4
Sorghum nitidum 26 10.3
Echinochloa colona (L.) Link 25 9.9
Brachiaria mutica L. 10 3.9
Miscellaneous 29 10.5
Miscellaneous weeds 93 27.0
Monochoria vaginalis (Bum. f.) Presl. 7 7.5
Ipomoea aquatica Forsk. 18 19.4
Marsilea crenata Presl. 14 15.0
lschaemum rugosum Salisb. 46 49.5
Miscellaneous 8 8.6
and Echinochloa spp. were cited by 34% and 75% of the farmers. respectively. Grassy
weeds need the same habitat as young rice plants: therefore, Leptochloa and
Echinochloa were dominant. The dominance of grassy weeds is influenced by soil
conditions, particularly less soil moisture. On the other hand, weeds with broad leaves
such as Monochoria vaginalis need a flooded condition, which occurs a few weeks
after seeding, thus causing no problems to the rice plants at their early stage.
Sources of rice weeds. According to 49% of the farmers, weeds come from the
soil, followed by 26% who reported that weeds are mixed with rice seeds, and 13%
who indicated that they are brought in by irrigation water. Other farmers specified
that weed seeds are blown in by the wind (13%), mixed with animal manure (2%).
scattered by cattle (1 %), and come from a neighbor’s field (1%). As a seed bank, the
soil is the main source of weed seeds because every season, before or after rice is
harvested, shedding weed seeds are buried in the soil by plowing (Mortimer 1994). At
harvest time, seeds of L. chinensis, E. crus-galli, M. vaginalis, and Echinochloa spp.
get mixed with rice seeds because their ripening period occurs at about the same time.
Farmers' perceptions and practices in weed management in Thailand 193

Table 5. Herbicides applied by farmers (n=231) at various crop stages in Thailand
in 1996.
Herbicide
Stage
Seedbed Seedling Vegetative Panicle Reprolnitiation
ductive
Preemergence 6 (46.1%) a 113 (68.1%) 29 (35.4%) 5 (50%) 3 (50%)
Preplanting 2 (15.4%) 23 (13.8%) 27 (32.9%) 0 (0%) 0 (0%)
Postemergence 5 (38.5%) 30 (18.1%) 2 (2.4%) 0 (0%) 3 (50%)
No answer 0 (0%) 0 (0%) 2 (2.4%) 0 (0%) 3 (50%)
Total 13 (5.6%)' 166 (71.9%) 82 (35.5%) 10 (4.3%) 6 (2.6%)
a Column totals.
b Row totals.
Effect of weeds on rice. Asked what weeds do to the riceficld, 80% of the farmers
indicated that weeds reduce rice yield, followed by 74% who said that weeds lower
the quality and price of rice and 49% who stated that they serve as hosts for diseases
and rodents. Weeds reduce rice yield, an effect that coincided with past reports of
farmers to the government about Oryza rufipogon Griff., Ischaemum rugosum Salisb.,
Chara zeylanica Willd., and Fimbristylis miliacea (L.) Vahl having caused yield losses
in Prachin Buri, Phra Nakhon Si, Ayutthaya, Surat Thani, and Ubon Ratchathani provinces
(Vongsaroj 1991).
Weed control methods. To determine weed control methods that are familiar to
farmers, they were asked how they could reduce weed problems in their ricefields.
The chemical method or herbicide application was the option known to 77% of the
farmers, followed by the mechanical method such as plowing or thorough land preparation,
which 53% of them mentioned. Other methods reported were flooding the
field (0.7%) and applying salt on weeds (0.3%).
Herbicides applied. Overall, 71.9% of the farmers made herbicide applications
at the seedling stage, followed by the vegetative stage (35.5%) (Table 5). Few applied
herbicides at the seedbed stage (5.6%), panicle initiation (4.3%), or reproductive stage
(2.6%). Within the seedbed stage, most of the herbicides applied were preemergence
(46.1%), followed by postemergence (38.5%). At the seedling stage, 68.1% used
preemergence herbicides. At the vegetative stage. the types of herbicide used were
somewhat evenly distributed among preemergence (35.4%), preplanting (32.9%), and
postemergence (29.3%).
Weeding operation
Timing of weeding. According to 58% of the farmers, weeding was carried out during
the vegetative stage, followed by 23% who performed weeding at the seedling stage
and 17% who did so at panicle initiation. A few did so before the seedling stage (3%)
and at the reproductive stage (9%).
Persons responsible for weeding. Some 80% of the respondents said that they
were responsible for weeding in their ricefields, 31% reported that their spouse per-
194 Meenakanit and Vongsarol

formed weeding, and 11% specified hired labor. A few (7%) cited other persons such
as children or neighbors.
Cost of weeding. Exactly 80% of the farmers reported spending from

Table 6. Water management procedures applied by farmers (n=291) in the paddy at various
crop stages in Thailand in 1996.
Stage
Seedbed Seedling Vegetative Panicle Reproductive
initiation
Flooded field with 35 (77.8%) b 247 (84.9%) 156 (53.6%) 113 (38.8%) 24 (8.2%)
1-40 cm water
Increase water 7 (15.6%) 9 (3.1%) 7 (2.4%) 7 (2.4%) 6 (2.1%)
level
Don’t know 3 (6.7%) 1(0.3%) 0 (0.0%) 0 (0.0%) 3 (1.0%)
Total c 45 (15.5%) 257 (88.3%) 163 (56%) 120 (41.2%) 33 (11.3%)
a Multiple responses.
b Column totals.
c Row totals.
Table 7. Farmers' attitudes toward weeds and weed management in Thailand in
1996.
Attitude statements
Agree No opinion Disagree
No. % No. % No. %
Early detection of weeds is 293 97.7 0 0 5 1.7
important.
Weed control involves a lot of 219 73.0 9 3.0 68 22.7
hard work and is not beneficial.
Seed cleaning is a waste of 196 65.3 14 4.7 86 28.7
time.
Good land preparation will 280 93.3 1 0.3 16 5.3
control weeds.
Synchronized planting can help 205 68.3 15 5.0 74 24.7
reduce the weed population.
Hand weeding is a waste of 233 77.7 6 2.0 58 19.3
time.
Applying herbicides at the 272 90.7 4 1.3 15 5.0
recommended dosage is
important.
Direct seeding or broadcasting 161 53.7 5 1.7 52 17.3
can increase the weed population.
Herbicides cannot control the
147 49.0 17 5.7 121 40.3
weed problems in my field.
A modern, progressive farmer 144 38.0 12 4.0 172 57.3
should use a lot of herbicides
for weed control.
To attain higher yields, my 295 98.4 1 0.3 1 0.3
crop should be free from weeds.
resorted to hired labor (6.7%), children (2.7%), parents (2.3%), neighbors (0.3%),
and an irrigation officer (0.3%).
Water management cost from $400 per season. Results showed that 59%
of the farmers reported spending between

Attitudes toward weeds
To determine farmers’ attitudes toward weeds and weed management, they were asked
to express their agreement or disagreement with 11 attitude statements (Table 7).
Based on respondents’ choices in 5 of the 11 attitude statements on key issues in weed
management, it appears that they have the proper attitudes toward weed management.
As shown in Table 7, 97.7% of the farmers agreed that early detection of weeds is
important, good land preparation will control weeds (93.3%), synchronized planting
can help reduce the weed population (68.3%), applying herbicides at the recommended
dosage is important (90.7%), and, to attain higher yields, the rice crop should be free
from weeds (98.4%).
On the other hand, 73% of the farmers believed that weed control involves a lot
of hard work and is not beneficial. that seed cleaning is a waste of time (65.3%). and
that hand weeding is a waste of time (77.7%). Respondents expressed some ambivalence
about the other attitude statements (Table 7).
Discussion
Most of the weeds reported by farmers are grasses and sedges. Infestation of grassy
weeds—namely, Echinocloa crus-galli and Leptochloa chinensis— cause a high yield
reduction, whereas sedges, such as Fimbrystylis miliacea and Cyperus difformis. bring
about lodging, which likewise results in a yield reduction. Weeds originate from many
sources: from the soil, from contaminated rice seeds, and from irrigation water. According
to scientists, the weed seed bank builds up from the constant shedding of
weed seeds.
Most farmers in Thailand use herbicides. The size of the riceland being cultivated
and the effectiveness, relatively lower cost, and availability of herbicides, compared
with other weed control methods such as increased seeding rate and water management,
might have contributed to the dependence on herbicides in direct-seeded
areas. In addition, the promotion of herbicides in boxing programs on television has
directly encouraged use, whereas farmer training programs might have unconsciously
promoted herbicide use when encouraging farmers to shift to direct seeding.
Farmers’ knowledge gaps on the importance of weed identification and the relationship
to timing and choice of appropriate herbicides for effective weed management
open up an intervention opportunity in weed management. In particular, farmer
training programs should emphasize the relationships among weed identification, water
management, and the appropriate herbicides to use at critical crop stages.
The importance of land leveling presents another intervention point for Thai farmers.
In Malaysia, one of the key factors in effective weed management was land leveling.
Ho (1994) reported how a strategic extension campaign to motivate and educate
farmers on the proper use of herbicides and nonchemical methods succeeded in bringing
about awareness that weeds are manageable and land leveling is important. As a result
of the extension campaign, Echinochloa infestations were reduced by 66% and
farmers’ use of correct herbicide dosages increased by 21%. The farmers’ practice of
land leveling increased markedly, from 17% before the campaign to 89% after it,
Farmers’ perceptions and practices in weed management in Thailand 197

which might have accounted for the increase in herbicide effectiveness (Adhikarya
1994). In Thailand, future training programs should emphasize land leveling to facilitate
water management and reduce weed infestations. With good land leveling, water
can be used to control weeds effectively.
Thai farmers seem to have the proper attitudes in weed management; this should
be considered a strength and it opens the way for future intervention opportunities.
But although Thai farmers realize the importance of weed control, their attitudes toward
carrying it out are not consistent; many consider it a waste ot time. Perhaps this
has to do with their knowledge gap on how efficient and effective weed management
should be carried out.
Some farmers consider the rice variety as a means of minimizing, if not totally
controlling, weeds in the field. Tall varieties are popular because of their market value,
their larger leaves, and their horizontal root distribution, which promotes competition
with weeds. The hybrid variety has these characteristics, but the stem is shorter, the
leaves are more erect, and the root is vertical, which makes the hybrid inferior in
competition with weeds. It is also more responsive to fertilizer than the local variety
and can be grown throughout the year.
The size of the rice farm was found to play an important role in farmers’ weed
management practices. On smaller farms, where manual methods are usually used,
weed control is often not done effectively because farmers tend to wait before controlling
any leftover weed species. In contrast, larger farms tend to achieve more efficient
weed control because of herbicide applications.
References
Adhikarya R. 1994. Strategic extension campaign: a participatory-oriented method of agriculrural
extension. Rome (Italy): FAO.
Bhandhufufalk A, Hare CJ. 1985. Sofit 300 EC: practical considerations and benefits from its
use in wet sown rice in Thailand. Proc. Asian Pacific Weed Sci. Soc. Conf. 10(1): 168-178.
Chinawong S. 1992. Effect of water level on the efficacy of several herbicides to barnyard
grass. Weed Res. (Japan) 37(3):248-250.
Chinawong S. 1993. Combined application of herbicides for weed control in wet-seeded rice in
Thailand. Japan J. Trop. Agric. 73(3):305-307.
Choutummatut S. Hongtrakul V, Thongdeethae P, Kwasard P. Junhuatong S, Makdee K. 1994.
Effect of herbicides at different stages of Echinochloa spp. Annual report. Bangkok (Thailand):
Pathum Thani Rice Research Center, Rice Research Institute. Department of Agriculture.
Day BE. 1972. Nonchemical weed control. In: Pest control strategies for the future. Washington
(D.C., USA): National Academy of Sciences.
De Datta SK. 1990. Technology and economics of weed control in small rice farms in the Asian
tropics. In: Weed problems and their economic management. Proceedings of the 12th
Asian-Pacific Weed Science Society and Korean Society of Weed Science, Seoul, Korea.
p. 23-37.
Denning GL, Jayasuriya SK, Huey BA. 1983. Constraints to the adoption of new weed control
technology in rice. In: Weed control in rice. Los Baños (Philippines): IRRI. p 345-360.
198 Meenakanit and Vongsaroj

DOA (Department of Agriculture). 1989. Weed control handbook. Botany and Weed Science
Division. 66 p.
DOAE (Department of Agricultural Extension). 1995. Weed control group. Plant Protection
Division. 15 p.
DOAE (Department of Agricultural Extension). 1994. Agricultural statistics (2537). Planning
Division, Bangkok (Thailand): DOAE.
Heong KL. Teng PS, Moody K. 1995a. Managing rice pests with less chemicals. GeoJournal
35(3):337-349.
Heong KL, Escalada, MM, Lazaro. AA. 1995b. Misuse of pesticides among rice farmers in
Leyte, Philippines. In: Pingali PL. Roger PA. editors. Impact of pesticides on farmers'
health and the rice environment. Noewell (Mass., USA): Kluwer Academic Publishers
and Manila (Philippines): International Rice Research Institute. p 97-108.
Ho NK. 1991. Integrated weed management of rice in Malaysia: some aspects of the Muda
Irrigation Scheme. Paper presented at the FAO-CAB International Workshop on Appropriate
Weed Control in Southeast Asia. 17-18 May 1994, Kuala Lumpur, Malaysia.
Jirström M. 1996. In the wake of the Green Revolution: environment and socio-economic
consequences of intensive rice agriculture — the problems of weeds in Muda, Malaysia.
Sweden: Lund University Press.
Kanchanomai P. 1981, Pre-germinated direct-seeded rice—new methodof rice cultivation in
Thailand. Department of Agriculture, Ministry of Agriculture and Cooperatives. 41 p.
Kanchanomai P, Vongsaroj P, Kongkanond A, Supapoj N, Chethul C, Chaimanit L. 1970. Comparisons
on performance of direct-seeded rice and transplanted rice under chemical control
treatment. In: Annual report. Technicial Division. Department of Agriculture. p 911-
918.
Labrada R, Parker C. 1994. Weed control in the context of integrated pest management. In
Weed management for developing countries. FAO Plant Production and Protection Paper
120. Rome: FAO. p 1-8.
Meenakanit L, Tanaseth A, Nuchporn S. 1994. Report on eradication campaign on Echinochloa
spp. and Leptochloa pp. Paper presented at the National Weed Science Conference. Kon
Kaen, Thailand. p 3-10.
Moody K. 1993. Weed control in wet-seeded rice in tropical Asia. Extension Bulletin No. 364.
Taiwan: ASPAC Food & Fertilizer Technology Center.
Mortimer AM. 1994. The classification and ecology of weeds. In: Labrada R. Caseley JC,
Parker C, editors. Weed management for developing countries. FAO Plant Production and
Protection Paper 120. Rome: FAO. p 11-26,
Panitchpat W. 1994. Improved rice direct seeding in Thailand. Paper presented at the Japan
Research Centre for Agricultural Science (JARCAS) Conference on Rice Direct Seeding
Culture, 4-5 Feb 1994, Tsukuba, Japan.
Vongsaroj P. 1991. Research and development for weed control in rice. In: Technical Bulletin,
Department of Agriculture, Ministry of Agriculture and Cooperatives.
Notes
Authors’ addresses: L. Meenakanit, Department of Agricultural Extension (DOAE), Bangkok,
Thailand 10900; P. Vongsaroj, Department of Agriculture (DOA), Bangkok, Thailand
10900.
Citation: Heong, KL. Escalada MM, editors. 1997. Pest management of rice farmers in Asia.
Manila (Philippines): International Rice Research Institute.
Farmers' perceptions and practices in weed management in Thailand 199

CHAPTER 14
The changing role of women
in rice pest management
in central Thailand
L. Meenakanit, M.M. Escalada, and K.L. Heong
We conducted a survey of 221 women farmers in central Thailand to
assess the role of women in rice pest management, analyze associated
factors, and determine training needs to improve pest management
knowledge and practices. Most of the women farmers were
31-40 years old, married, had a primary education, and were involved
in all rice production operations. To enable them to engage in
rice farming, most women had to give up housework and other domestic
responsibilities.
Most of the women farmers interviewed perceived that the brown
planthopper was the most serious insect pest. To control pests,
most used pesticides and applied their first sprays within the first
30 days after planting. Pesticide application frequencies ranged from
1 to 10 per season, with 1-3 times as the most common. Decisions
on the amount of money for pesticides were made by the woman or
a male household member. About a third of the women farmers interviewed
reported applying pesticides themselves. Most were aware
of pesticide health hazards and more than half reported that they
sometimes experienced illness after spraying. About half had attended
some form of pest management training and among the untrained
women farmers, most expressed a lack of interest in attending training
because they had no time. Most believed that pesticide applications
would increase rice yields.
Introduction
Women play a major role in food production in most developing countries but their
contributions as household food providers have remained less visible. In Thailand,
the proportion of female farmers in the total agricultural labor force is reported to be
60%, the highest in Southeast Asia. In the past few years, women in rice-farming
communities in central Thailand have assumed increasing responsibilities in rice production
as men abandon agriculture to seek better-paying jobs in urban areas. The
growing male out-migration from rural areas has forced women to take over tasks
201

traditionally performed by men (Dulyapach 1991, FAO 1985, Tacio 1996). Although
commonly excluded from the mainstream of agricultural development, women inevitably
perform spraying tasks that expose them to toxic pesticides. With limited access
to information and training opportunities on safe pesticide use, women farmers become
particularly vulnerable to hazardous chemicals.
This rapid change in the role of women in rice production has occurred largely
because of the shortage of labor in rural areas, caused by urban pull factors such as
higher wages in the city and in the international job market (Dulyapach 1991). Migration
data show that male migrants moved to Bangkok during the slack agricultural
season primarily to look for work (ESCAP 1988). About three-fourths of the employed
in-migrants to the urban centers were reported to be farmers and farm workers
before they moved (NSO 1988). Most of them came from the central region (NSO
1984). The seasonal nature of agricultural work in Thailand engenders a complex
pattern of movements between areas. The recent National Migration Survey reported
that the central region has the highest level of gross movement, which is perhaps
influenced by its proximity to Bangkok. The central region competed with Bangkok
in the number of interregional out-migrants. Most migrants from Bangkok and the
central region went to the northeast and the north, although Bangkok was also an
important destination for migrants from the central region (Chamratrithirong et al
1995).
As the Thai economy surges forward, this trend is likely to worsen. Questions on
the effects of urban migration of male farm labor on women’s role in pest management
and appropriate training needed to improve their skills should be addressed.
The Department of Agricultural Extension (DOAE) planned to organize a special
integrated pest management (IPM) training course to acquaint women with pest
management concepts. To assess the key issues affecting women’s participation in
rice pest management, a diagnostic survey using the rapid rural appraisal (RRA) approach,
followed by a formal survey, was carried out in central Thailand between
March and June 1992. The objectives of the surveys were (1) to assess changes in the
role of women in rice pest management, and (2) to analyze associated factors and
determine training needs to improve pest management knowledge and practices of
women farmers.
This paper reports the results of the survey conducted in Chainat and Lop Buri
provinces, central Thailand; observations from the RRA interviews appear in the discussion.
Methods
Diagnostic interviews using the RRA approach were first conducted to identify the
relevant issues in women’s involvement in rice pest management. Results of these
interviews were used as inputs in developing the questionnaire for the survey. The
survey questionnaire was translated into Thai and pretested with a small group of rice
farmers to determine the clarity and relevance of the questions. After considering the
feedback from the pretest, we revised the questionnaire.
202 Meenakanit et al

To collect the data, personal interviews using the revised questionnaire were conducted
by eight graduate students from Kasetsart University, who were closely supervised
by the authors. Prior to field work, the students were trained on the objectives of
the survey, interviewing techniques, and the use of the questionnaire. The survey was
conducted in May 1992.
Survey areas
Geographically, Central Thailand is composed of several river basins collectively called
the Chao Phraya basin. The plains are mostly covered by brackish alluvial and riverine
alluvial soils. The eastern and western sections of the plains contain noncalcic
brown and low humic gley soils. Mountains also border the eastern and western edges
(Takaya 1987). The region lies in the tropical monsoon belt, which is influenced by
the southwest monsoon, which brings rain from May to October, and the northeast
monsoon, which brings the dry season the rest of the year. The annual temperature
averages around 28 °C.
Chainat Province in central Thailand (Fig. 1) covers a total area of about 263,600
ha, of which 180,800 ha are devoted to agriculture. In 1992, the farming population
was estimated at 221,200, which represents about 63.4% of the total population of
348,600. Farm households have been estimated at 48,800, distributed in 444 villages
in 5 1 districts (DOAE 1992).
Lop Buri (Fig. l), on the other hand, spans 619,840 ha. Of this, arable land constitutes
346,560 ha and irrigated areas 97,500 ha. Rice, maize, sorghum, beans, and
sugarcane are crops grown by farmers. The farming population is about 348,200 in
67,300 households.
Respondents
The random sample of 221 women farmers was selected from five districts in two
provinces: Hankha (15.8%), Sankhaburi (24.4%), Manorom (11.3%), Amphur Muang
(29.4%), and Amphur Banmoh (19%). These rice-growing districts were selected randomly
from the two provinces and, in each district, only women who had been engaged
in rice farming were selected for the survey.
Data analysis
The survey data obtained were coded and entered into a spreadsheet program using a
microcomputer. After validation, the data file was uploaded to the mainframe IBM4361
at the International Rice Research Institute. Frequency tables were generated using
the FREQ procedure available in Statistical Analysis System (SAS 1985). Percentages
were based on the number of respondents rather than the total sample. In cases
where multiple responses were obtained, the total sample size was used.
The changing role of women in rice pest management in central Thailand 203

Fig. 1. Location of Chainat and Lop Buri provinces in central Thailand.
204 Meenakanit et al

Results
Demographic profile
About a third of the women farmers (35.3%) were 31–40 yr old, followed by 23.5%
in the 41–50 age group. More than a fifth (20.8%) were less than 30 yr old. Nine out
of 10 women farmers interviewed were married; the rest were either single (7.7%) or
widows (2.3%). The respondents’ education was low. More than four-fifths (88.2%)
of the women had only a primary level of education, 5% reached the secondary level,
and a negligible 0.4% went to college. Some 6.4% admitted that they had not received
any schooling.
Women's involvement in rice-farming activities
Women were involved in all rice production operations. Overall, women in the study
areas tended to perform more rice-farming tasks (40.7%) than the males in the households,
that is, their husbands or brothers (38.4%) and hired labor (1 8.5%). Activities
such as selecting rice varieties, planting, applying fertilizer, weeding, field checking,
drying, and marketing were dominated by women. In households where male labor
was not available, women also performed jobs that required physical strength, such as
land preparation using a hand-operated tractor and pesticide application using a knapsack
sprayer (Table 1).
In general, respondents considered land preparation (27%) and pesticide application
(20%) to be the most demanding farm activities. The response pattern varied
between the two provinces, with more women farmers in Lop Buri (38%) than in
Chainat (14%) finding land preparation more demanding. The pattern was reversed
for pesticide application, with more Chainat respondents (36%) than their Lop Buri
counterparts (6%) regarding this as a demanding task.
Table 1. Persons who performed each rice-farming operation
in the respondents’ households. Percentages were based on
the total number of respondents.
Self/ Husband/ Hired Other
Farm activity woman brother laborer
(%)
Selection of rice varieties 54.9 42.0 0.0 3.1
Land preparation 34.7 40.1 22.9 2.3
Planting 45.4 41.5 11.5 1.5
Fertilizer application 47.1 45.4 5.8 1.7
Weeding 39.4 43.0 14.3 3.3
Field checking 49.5 46.4 1.2 2.8
Pesticide application 26.9 50.0 19.9 3.2
Harvesting 26.5 23.4 47.7 2.4
Threshing 13.4 13.8 70.1 2.8
Drying 51.8 38.9 8.3 1.0
Marketing 57.6 37.1 2.8 2.5
Overall % 40.7 38.4 18.5 2.4
The changing role of women in rice pest management in central Thilland 205

To enable them to engage in rice farming, most women had to give up housework
(33.3%), vegetable growing (6.7%), and livestock raising (6.7%), although 30% said
that they continued to perform their usual domestic responsibilities along with their
rice-farming tasks.
Cropping patterns
Varieties planted. There appears to be a sharp contrast between the two provinces in
rice varieties planted by women farmers. In Chainat, more farmers reported planting
RD23 (34%) and Suphan Buri 90 (22%), whereas in Lop Bun, 57% of the farmers
cultivated local varieties, followed by 17% who grew B4 and 8% who grew RD23.
Crop establishment. More than four-fifths (84%) of the women farmers directseeded
their crops versus only 14% who transplanted and 3% who did both. For the
two provinces, all Chainat farmers direct-seeded their rice crop, whereas Lop Buri
farmers used different crop establishment methods: 66% practiced direct seeding,
28% did transplanting, and 6% did both.
Cropping patterns. Slightly more respondents (48%) planted rice followed by a
fallow than those who grew two rice crops a year (44%). Only a few (7%) cultivated
other crops after rice.
Pest management perceptions and practices
Most respondents perceived hoppers to be the most serious pests (61.5%), followed
by rats (10.9%), diseases (7.7%), thrips (5%), stem borers (3.6%), and lepidopterous
leaf feeders (2.3%). In Chainat, 74.6% of the respondents reported hoppers as the
main pest, whereas in Lop Buri 48.1% mentioned them. More Chainat farmers reported
diseases than their Lop Buri counterparts (Table 2).
Table 2. Most important pests reported by
women farmers in central Thailand in 1992.
Province
Pests Chainat Lop Buri Overall
(%)
Hoppers
Stem borer
Lepidopterous
leaf feeders
Rice bug
Thrips
Rats
Weeds
Fungi
Blast
Diseases
Others
Don’t know
None
74.6 48.1 61.5
0.9 4.7 3.6
0.9 3.7 2.3
1.8 0.0 0.9
0.9 9.4 5.0
0.9 21.7 10.9
0.0 2.8 1.4
8.8 0.0 4.5
2.6 0.0 1.4
3.5 0.0 1.8
1.8 0.9 1.4
0.9 3.6 4.5
0.0 0.4 0.4
206 Meenakanit et al

To control the major pests, women farmers in Chainat and Lop Bun reported that
they applied pesticides (67%), used baits or poison to kill rats (10%), used a light trap
or black light (4%), drained the fields (3%), or consulted their extension technician
before taking any control action (2%).
Pesticide application patterns
In both provinces, 90% of the women farmers reported applying pesticides in the
1992 wet season. These include insecticides, rodenticides, fungicides, and herbicides.
In Chainat, 92% of the farmers used pesticides, versus 87% in Lop Bun.
More than three-quarters (75.8%) of the farmers in both provinces applied their
first insecticide spray in the first 30 d after planting. Another 18.8% applied it in the
second month. For timing of application by crop stage, respondents applied pesticides
in the seedling (44.7%) and tillering (28.7%) stages, although more Chainat
respondents (81.6%) than their Lop Bun counterparts (54.2%) used pesticides in the
seedling stage (Table 3).
Pesticide application frequencies of farmers ranged from 1 to 10 per season, with
1–3 times as the most common (Table 3). Slightly more Chainat respondents were
heavy pesticide users than those in Lop Buri. Women farmers used pesticides, par-
Table 3. Pesticide application patterns of women farmers in central
Thailand, 1992.
Province
Pattern Chainat Lop Burl Overall
Timing of first insecticide
application (wk after planting)
1–4 83.3 67.8 75.8
5–8 6.3 32.2 18.8
>8 10.4 0.0 5.4
Frequency of pesticide application
in a season
1 36.4 34.4 35.4
2 26.3 37.6 31.8
3 20.2 23.7 21.9
4 7.1 2.1 4.7
5 2.0 2.1 1.0
6 3.0 2.1 2.6
7 2.0 0.0 1.0
10 3.0 0.0 1.6
Timing of pesticide application
(by crop stage)
Seedling 81.6 54.2 44.7
Tillering 31.6 57.0 28.7
Booting 30.7 7.5 12.7
Heading 21.9 0.9 7.7
Flowering 6.1 0.9 2.4
Milking 3.5 5.6 3.0
Ripening 1.8 0.9 0.9
(%)
The changing role of women in rice pest management in central Thailand 207

ticularly insecticides, either to prevent or control pest infestations. Some 55% of the
respondents indicated that they had sprayed for control rather than prevention (34%).
At the seedling stage, insecticides were applied more for prevention than at the tillering,
booting, heading, and flowering stages, when insecticides were used for control.
Some 64% of the farmers estimated that insecticides could kill 75–100% of the
pests in their ricefields, whereas 27% judged that only 50% of the pests were killed. A
few believed that fewer than 50% of the pests were eliminated after pesticide applications.
Involvement in decision making and purchase and use of pesticides
Decisions on the amount of money to be spent for pesticides were made by either the
woman (30%) or a male household member, the husband or brother (37%). Fewer
than a third of the respondents (28%) stated that both the woman and the husband
decided jointly on allocations for pesticides. Slight differences in decision-making
patterns were found between the two provinces, with more Chainat women farmers
(34%) reporting that they made the decisions on the amount of money to be used for
pesticides.
Fewer than half of the farmers (46%) indicated that women bought pesticides,
whereas 38% said that the husband or brother purchased them. Only 10% of the farmers
said that both the woman and the husband bought the pesticides. The two provinces
showed some differences. Chainat women farmers appear to have more participation
(52%) than their Lop Buri counterparts (39%) in the purchase of pesticides.
About a third of the women farmers (33%) applied pesticides in their own ricefield.
More women farmers in Lop Buri (36%) did their own spraying than in Chainat
(30%). When asked who applied pesticides the year before the survey (1991), women
farmers mentioned their husband or brother (80%), father (5%), and hired labor (5%).
Among the main reasons given by women farmers for applying the pesticides themselves
were unavailability of hired labor (47%), lack of male family labor (26%), and
lack of capital to pay hired labor (5%).
Women farmers were usually aware of pesticide health hazards. More than half
of the respondents (53%) reported that they got sick or experienced such symptoms
as dizziness, dryness of throat, tiredness, headaches, and vomiting after pesticide applications.
Some 47% of the women farmers acknowledged that doing their own field
spraying exposed them to toxic chemicals.
Ownership and access to sprayers
More than two-thirds of the farmers (67.4%) had their own sprayers whereas 30.8%
did not own any. More Lop Buri farmers (70%) owned sprayers than in Chainat (65%).
Many farmers owned sprayers that had less than 2 gallons’ capacity (41%). Some
30% owned knapsack sprayers and 12% had mist blowers.
Pesticide safety
More than a third of the women (39%) sold the used pesticide containers, whereas
39% left them in the field and 16% buried them. More women farmers in Chainat sold
208 Meenakanit et al

them (71%) than those from Lop Buri (13%). On the other hand, more Lop Buri
farmers (72%) left the pesticide containers in the field.
The women who applied pesticides protected themselves by covering their mouth
and nose with a piece of cloth (24%) and wearing long-sleeved shirts (18%), and by
taking a bath (17%) and washing their hands (16%) after spraying. Few women (2.6%)
wore a mask while spraying.
Pest management advice and training
Information sources on pest management consulted by women farmers included extension
officers, or kaset tambon (39%), plant protection officers (22%), neighbors
(18%), and pesticide sales agents (11 %). Others mentioned training, radio, and television.
Farmers who cited multiple sources of advice on pest management were further
asked to name which sources they believed were the most credible, and the agricultural
extension officers (43%) and plant protection officers (21%) were mentioned
most.
Fewer than half of the women (41%) had not attended any pest management
training. Among the untrained women farmers, most (92%) indicated a lack of interest
in being trained. Even on a per province basis, women farmers had consistent
responses. Some 72% of the women said they had no time to attend, 11% replied that
they couldn’t leave home, and 13% admitted that they were not interested. Other
respondents were undecided about attendance at training, believed they had sufficient
knowledge, or would rather have their husbands attend the training.
Pest management knowledge
More than two-thirds of the women farmers (68%) thought that pesticide applications
would increase rice yields, whereas 23% disagreed and 9% had no opinion. More
women farmers in Lop Buri (79%) believed that pesticides would increase yields than
in Chainat.
Perceptions of leaf-feeding insects
Grasshoppers (38.5%), leaf rollers (13.6%), leaffolder (11.5%), and lepidopterous
larvae (9.4%) were the most common leaf feeders known to farmers. Women farmers
mentioned other leaf feeders such as cutworms (5.2%), stem borers (4.9%), thrips
(2.4%), and brown planthopper (1.4%).
Nearly half of the women farmers (49.8%) did not believe that leaf-feeding insects
would cause severe damage to rice. More farmers in Chainat (64%) than in Lop
Buri (35%) believed that leaf-feeding insects would not cause severe damage to the
rice crop.
More than half of the farmers (51.4%) indicated that leaf-feeding insects would
not cause yield loss, whereas 34.5% believed otherwise and 14.1% had no opinion
(Table 4).
Nearly half of the farmers (46.6%) reported that spraying chemicals to control
leaf feeders was unnecessary. Twice as many Chainat farmers (61.4%) as their Lop
Bun counterparts (30.8%) believed that there was no need to spray insecticides for
The changing role of women in rice pest management in central Thailand 209

Table 4. Women farmers’ perceptions of leaf-feeding insects.
Province
Response Chainat Lop Buri Overall
(%)
Leaf-feeding insects cause severe
damage to the rice crop
Agree 28.1
Disagree 64.0
No opinion 7.9
Leaf-feeding insects do not cause
yield loss
Agree 57.9
Disagree 31.6
No opinion 10.5
Need to apply chemicals to control
leaf feeders
Agree 29.8
Disagree 61.4
No opinion 8.8
Need to spray chemicals early to
control leaf feeders
Agree 24.6
Disagree 64.0
No opinion 11.4
55.1 41.2
34.6 49.8
49.8 9.0
44.9 51.4
37.4 34.5
17.8 14.1
57.9 43.4
30.8 46.6
11.2 10.0
48.6 36.2
34.6 49.8
16.8 14.0
leaf-feeding insects. In contrast, more Lop Buri farmers (57.9%) thought that insecticide
applications for leaf feeders were necessary (Table 4). Likewise, 36.2% of the
farmers believed that chemical control for leaf-feeding insects had to be done early.
More farmers in Lop Buri (48.6%) than in Chainat (24.6%) indicated that chemical
control for leaf feeders had to be done early in the crop season.
Discussion
The lack of farm labor seemed to have forced women farmers in central Thailand into
taking on more tasks in rice farming than before. Pest management, traditionally a
man's activity, is one of these. The survey found that women generally had poorer
knowledge of and lacked basic skills in pest and disease diagnosis, pesticides, application
methods, and concepts of natural biological control. Although they were aware
of the acute toxicity of pesticides, they had little knowledge of the chronic effects.
Wherever possible, the women farmers had tried employing labor for sprays. But
with further male migration to cities and higher labor costs, more women may have
had to do their own spraying. This could account for the high percentage of hand
sprayers of lower capacities that were found in central Thailand. In Malaysia, where
spraying had been carried out predominantly by men, farmers usually owned knapsack
sprayers (Heong et al 1992).
210 Meenahanit et al

Women’s attitudes and practices toward pest management appeared rather similar
to those of farmers elsewhere. The descriptive profile generally fits into that described
by Lim and Heong (1984). A large proportion of the women farmers perceived
that pesticides increase yields and that visible pests, such as leaf feeders, are
important. Most farmers apply insecticides during the first four weeks after crop establishment.
Similar trends in pest management practices were described in the Philippines,
Vietnam (Heong et al 1994), Malaysia (Normiyah et al 1995), and China
(this volume). These practices often result in high misuse (Heong et al 1995).
Because pest management appears to be a recently acquired responsibility of
women, opportunities may exist to incorporate the appropriate pest management concepts
at this formative stage. With training, the appropriate attitudes and beliefs in
pest management could be developed and unnecessary spraying prevented. IPM implementation
programs could focus on optimizing this opportunity.
Without training, women could be easily persuaded by aggressive advertising as
in detergent sales. Because the primary objective of advertising is to increase sales,
advertising pesticides would tend to increase their use. Women could become lockedin
to this need to either cure or prevent pests and diseases.
Perceptions, beliefs, and attitudes that affect farmer practices can be effectively
changed through participatory training methods. Although the training needs of men
and women farmers may be similar, several aspects could require special attention.
For instance, because they regarded training as time-consuming, most of the women
we interviewed were not willing to participate because of their household commitments.
Season-long training on IPM would be a burden to them. Farmer participatory
research methods to improve pest management perceptions and practices could be
useful in fitting training into women’s schedules. Farmer participatory research involves
encouraging farmers to engage in experimentation to enable them to adapt
new technologies and spread them to other farmers (Bunch 1989). In this approach,
participating farmers were encouraged to “test” a simple hypothesis, which was presented
as a heuristic (a rule) (Escalada and Heong 1993). Experimental results were
then shared with other members of the community.
References
Bunch R. 1989. Encouraging farmers’ experiments. In: Chambers R, Pacey A, Thrupp LA,
editors. Farmer first: farmer innovation and agricultural research. London: Intermediate
Technology Publications. p 55-61.
Chamratrithirong A, Archavanitkul K, Richter K, Guest P, Thongthai V, Boonchalaksi W,
Piriyathamwong N, Vong-Ek P. 1995. National migration survey of Thailand. Thailand:
Mahidol University at Salaya.
DOAE (Department of Agricultural Extension). 1992. Fruit crop for commercial and export
statistics for 1991-1992. 566 p.
Dulyapach P. 1991. A case study on the relationship between the integration of women in rural
development and women’s reproductive behavior in Thailand. Bangkok (Thailand): Department
of Agricultural Extension.
The changing role of women in rice pest management in central Thailand 211

Escalada MM, Heong KL. 1993. Communication and implementation of change in crop protection.
In: Chadwick DJ, Marsh J, editors. Crop protection and sustainable agriculture.
Chichester: John Wiley & Sons. p 191-202.
ESCAP (Economic and Social Commission for Asia and the Pacific). 1988. Trends in migration
and urbanization in selected ESCAP countries. Bangkok (Thailand): ESCAP.
FAO. 1985. Women in developing agriculture. Rome: Human Resources, Institutions and Agrarian
Reform Division, FAO.
Heong KL, Escalada MM, Lazaro AA. 1992. Early spraying by rice farmers in Leyte, Philippines.
Int. Rice Res. Newsl. 17(6):27-28.
Heong KL, Escalada MM, Lazaro AA. 1995. Misuse of pesticides among rice farmers in Leyte,
Philippines. In: Pingali PL, Roger PA, editors. Impact of pesticides on farmers’ health and
the rice environment. Norwell (Mass., USA): Kluwer Academic Publishers and IRRI. p
97-108.
Heong KL, Escalada MM, Vo Mai. 1994. An analysis of insecticide use in rice: case studies in
the Philippines and Vietnam. Int. J. Pest Manage. 40:173-178.
Lim GS, Heong KL. 1984. The role of insecticides in rice integrated pest management. In:
Proceedings of the FAO/IRRI Workshop on Judicious and Efficient Use of Insecticides on
Rice. Los Baños (Philippines): International Rice Research Institute.
Normiyah R, Chang PM, Azmi M, Aznan A. 1995. Pest management practices of rice farmers
in the Muda irrigation scheme, Malaysia. Makalah Sesekala Bil 14. Kuala Lumpur (Malaysia):
MARDI. 14 p.
NSO (National Statistical Office). 1984. Survey of migration into the Bangkok metropolis and
vicinity, 1984. Thailand: Office of the Prime Minister.
NSO (National Statistical Office). 1988. Survey of migration into the Bangkok metropolis.
Thailand: Office of the Prime Minister.
SAS (Statistical Analysis System). 1985. SAS user’s guide: statistics. Version 5 Edition. Cary
(N.C., USA): SAS Institute, Inc.
Tacio H. 1996. Women rule the farms, too. Today. 22 Oct 1996. p 15.
Takaya Y. 1987. Agricultural development of a tropical delta: a study of the Chao Phraya delta.
Honolulu: University of Hawaii Press.
Notes
Authors’ addresses: L. Meenakanit, Plant Protection Service Division, Department of Agricultural
Extension, Bangkok 10900, Thailand; M.M. Escalada, Department of Development
Communication, Visayas State College of Agriculture, Baybay, Leyte 6521-A, Philippines;
K.L. Heong, Entomology and Plant Pathology Division, International Rice Research
Institute, Los Baños, Laguna, Philippines.
Acknowledgments: The authors would like to thank the Swiss Agency for Development and
Cooperation, through the Rice IPM Network based at IRRI, for funding the farmer survey
and rapid rural appraisal from which we have drawn the contents for this paper; Thelma R.
Paris, for taking part in the diagnostic interviews and sharing with us her insights on
gender issues; Amor A. Lazaro, for processing the survey data; the plant protection staff at
the DOAE head office in central Thailand, for facilitating the fieldwork; and the Kasetsart
University students, for conducting the interviews.
Citation: Heong KL, Escalada MM, editors. 1997. Pest management of rice farmers in Asia.
Manila (Philippines): International Rice Research Institute.
212 Meenakanit et al

CHAPTER 15
Pest management perceptions
and practices of rice farmers
in Long An Province, Vietnam
Vo Mai, N.H. Huan, K.L. Heong, M.M. Escalada, and A.A. Lazaro
A survey of 633 rice farmers conducted in August 1994 in two districts,
Tan Tru and Tan Thanh, in Long An Province, showed that
brown planthoppers, sheath blight, and leaffolders were farmers’
most important rice pests. Farmers often sprayed pesticides to control
them. Particularly for leaf-feeding insects, insecticide sprays were
applied during the first month after seeding. These insects were
perceived to be very damaging because they can cause heavy yield
losses in rice. Early season insecticide sprays for their control were
thus believed to be necessary. Organophosphates, such as
methamidophos, monocrotophos, and methyl parathion, were the
most popular insecticides used to control leaf feeders. These insecticide
applications may have been unnecessary because rice plants
have the ability to compensate for leaf defoliation. Early applications
of broad-spectrum insecticides may cause ecological disruptions that
can favor brown planthopper development. Farmer participatory research,
strategic extension campaigns, and pest management training
to help farmers realize that these early insecticide sprays are not
beneficial should be increased.
Introduction
Rice intensification programs in the 1970s introduced insecticides as a necessary input
for high yields. Prophylactic insecticide applications were “packaged” into production
programs and farmers were encouraged to use them at regular intervals. Government
subsidies and agricultural loan schemes together with strong marketing and
advertising by the chemical industry (Escalada and Heong 1993) further supported
insecticide use (Conway and Pretty 1991).
In Vietnam, rice production between 1970 and 1980 remained low at around 11
million t yr -1 . With the introduction of policies to improve contractual arrangements,
referred to as Contract 100, in 1981, rice production increased by about 30% (Cuc
1995). In 1989, Resolution N10 was introduced and rice production increased further,
213

eaching 22 million t yr -1 . This marked the beginning of decollectivization, which
encouraged farmers to invest in inputs and labor (Pingali and Xuan 1992). Thus, the
use of chemical inputs such as fertilizers and pesticides increased. Stimulated by aggressive
advertising and marketing, insecticide use in 1992 was around 5 sprays season
-1 (Heong et al 1994) and about 80% of the farmers applied their first sprays within
the first 40 d after sowing.
This paper presents the results of a baseline survey conducted in August 1994 to
document farmers’ pest management perceptions and practices in two districts, Tan
Tru and Tan Thanh, in Long An Province.
Long An Province
Long An is a province of the Mekong Delta in the southern part of Vietnam. The
province extends from 105° 30’ to 106° 45’ east, and from 10° 08’ 30” to 11° 02’ 30”
north (Fig. 1). The total area is 4,355 km 2 and agriculture plays a dominant role in the
economy. Rice is the principal crop; it is cultivated on 363,000 ha, 80% of which are
irrigated. The total farmer household population is 210,000, 85% of which is engaged
in rice farming. Long An produces 1.16 million t yr -l of rice. Besides rice, watermelon,
groundnuts, and cash crops are also cultivated. In most areas, farmers grow
two crops a year; in some areas, three rice crops year -1 are grown. Modern highyielding
rice varieties are commonly planted in more than 80% of the rice areas. In
some districts, the aromatic traditional varieties are popular, particularly during the
wet season.
Fig. 1. Map of Long An Province, Vietnam.
214 Vo Mai et al

Methods
In August 1994, a farmer survey was conducted in Long An Province to determine
rice farmers’ pest management practices and perceptions. Before being used, a structured
questionnaire was pretested by local technicians through field interviews with
20 farmers. The pretest was intended to solicit farmers’ responses to the survey questions,
which were used to further refine the questions. For the survey, 633 rice farmers
were selected randomly from 16 villages in two districts, Tan Tru and Tan Thanh,
where rice was the main crop. These farmers were interviewed by the agricultural
technicians trained to use the pretested survey questionnaire. The field data were immediately
translated, coded, and entered in a spreadsheet program. Frequency distributions
and cross-tabulations were generated using the frequency procedure in the
Statistical Analysis System (SAS) package (SAS 1985). Means were compared using
the t-test procedure. In addition, the authors carried out several follow-up focus group
interviews to obtain further information using probing techniques.
Results
Profile of farmers
More than half (55%) of the farmer respondents interviewed in Tan Tru and Tan Thanh
were 31–50 yr old. Only a small percentage (6%) did not have any schooling, whereas
81% had from 1 to 9 yr of education (Table 1).
Modern high-yielding rice varieties such as IR9729, IRS0404, and IR49517 were
the most common cultivars planted by the farmers during the summer-autumn season
of 1994. Farmers either kept their own seeds (34%) or exchanged them with their
neighbors (46%). More than 80% of the farmers in the survey followed a rice-rice
Table 1. Profile of farmer respondents in Tan Tru and Tan Thanh districts of Long An Province.
Age group (yr)
Tan Tru
N %
Tan Thanh
N %
Education
(no. of yr in school)
Tan Tru
N %
Tan Thanh
N %
70 15 4.8
66 20.7
121 37.9
58 18.2
38 11.9
31 9.7
5 1.6
Did not attend
school
1–5
6–9
10–12
16 years
7 2.2
170 54.7
73 23.5
59 19.0
2 0.6
30 9.3
176 54.8
90 28.0
25 7.8
0 0.0
Farm size (ha)
Tan Tru
N %
Tan Thanh
N %
Method of crop
establishment
Tan Tru
N %
Tan Thanh
N %
< 0.5 164 52.6 37 11.5 Direct-seeded 304 97.4 308 97.1
>0.5–1 105 33.7 85 26.5 Transplanted 7 2.2 4 1.8
>1–2 37 11.9 127 39.6 Both 1 0.3 6 1.1
>2–3 4 1.3 49 15.3
>3–4 0 0.0 9 2.8
>4 2 0.6 14 4.4
Pest management perceptions and practices in Long An Province, Vietnam 215

Fig. 2. Distribution of farmers’ first insecticide applications,
summer-autumn, Long An, Vietnam.
cropping pattern and 97% direct-seeded their rice. Almost 90% of the farmers cultivated
ricelands of less than 2 ha and 97% said they owned the land they managed.
Pest management practices
Farmers in Long An Province used mainly pesticides to manage pest problems. Nearly
all farmers (99%) used pesticides during the summer-autumn 1994 season. The 633
farmers interviewed reported using a total of 3,033 sprays. Of these, about 70% were
insecticides, 28% fungicides, and 2% herbicides. Most farmers (80%) applied pesticides
2–5 times. About 84% applied their first insecticide treatment within the first
month after crop establishment (Fig. 2); their main targets during this period were
thrips, leaffolders, brown planthoppers (BPH), caseworms, and stem borers. These
were also common targets at the booting stage. Sheath blight and yellow leaf disease
were targets from the booting to soft dough stage, and rice bugs at the milking stage.
Pesticide use
Organophosphates were usually the most popular insecticides used for insect pest
control in rice. Methamidophos was the predominant compound used to control
leaffolders, caseworms, and stem borers (Table 2). It was also used on thrips and rice
bugs, although most farmers used methyl parathion for these pests. For brown
planthoppers, fenobucarb was the most common pesticide used, followed by buprofezin
and methamidophos; for sheath blight, farmers used the fungicide hexaconazole.
About 34% of the total pesticide sprays were compounds in WHO category I and
30% were in category II. The rest were WHO categories III (1%), V (30%), and unclassified
(5%).
During the 1994 season, 2,123 insecticide sprays were applied by the farmers we
interviewed. Most sprays targeted one pest, although some targeted two or three (Table
216 Vo Mai et al

2). Of the 2,564 intended targets, 33% were leaffolders, 28% BPH, 14% thrips, 10%
stem borers, 7% caseworms, 4% rice bugs, and the rest other insects. The lepidopterous
leaf feeders constituted 43% of the intended spray targets.
Farmers in the two districts used 849 fungicide sprays and, of the 932 spray
targets mentioned, 71% were sheath blight, 26% were yellow leaf disease, and 2%
were blast (Table 3). Although more is known about sheath blight and blast, the causal
agent of yellow leaf disease is still unknown.
Farmers ranked the brown planthopper (27%) as the most important pest problem,
followed by sheath blight (26%), leaffolders (16%), and yellow leaf disease (10%).
Most farmers applied pesticides for these pests (Table 4). Only 0.5% of the farmers
mentioned that they used other control measures such as water management, baiting,
and handpicking.
Table 3. Fungicides used by farmers for disease control in Long An, Vietnam.
Target rice pests, no. of targets a , and % b
WHO Sheath Yellow leaf Overall
Fungicides classification c blight disease Blast
N %
Antibiotic
Validamycin
Fthalide + kasugamycin
Inorganic
Copper oxychloride
Triazole
Hexaconazole
Propiconazole
Dicarboximide
lprodione
Benzimidazole
Benomyl
Carbendazim
Thiophanate-methyl
Phthalimide
Captan
Dithiolane
lsoprothiolane
Cyclopentane
Pencycuron
Dithiocaibamate
V
V
II
V
V
V
V
V
III
192 (28.9) 20 (8.1)
1 (0.1) 1 (0.4)
4 (0.6) 4 (1.6)
279 (42.0) 73 (29.7)
31 (4.7) 12 (4.9)
19 (2.8) 14 (5.7)
55 (8.3) 90 (36.6)
24 (3.6) 22 (8.9)
14 (2.1) 2 (0.8)
3 (0.4) 1 (0.4)
16 (2.4) 4 (1.6)
3 (14.2) 215 23.1
1 (4.8) 3 0.3
0 (0.0) 8 0.9
1 (4.8) 353 37.9
0 (0.0) 43 4.6
0 (0.0) 33 3.5
0 (0.0) 145 15.6
0 (0.0) 46 4.9
1 (4.8) 17 1.8
0 (0.0) 4 0.4
10 (47.6) 30 3.2
V 3 (0.4) 2 (0.8) 0 (0.0) 5 0.5
Mancozeb V 1(0.1) 0 (0.0) 0 (0.0) 1 0.1
Organophosphorus
lprobenfos III 6 (0.9) 1 (0.4) 4 (19.0) 11 1.1
Arsenic
MAFA d 17 (2.6) 0 (0.0) 1 (4.8) 18 1.9
Total 665 (100) 246 (100) 21 (100) 932
a List of common names by chemical family. Many farmers applied fungicides for more than one disease problem.
b Percentages were based on column totals. c Il = moderately hazardous; III =slightly hazardous, V = unlikely
to present acute hazard in normal use. d Ferric ammonium salt of methane arsenic acid.
Source: CIRAD 1991.
218 Vo Mai et al

Table 4. Most important rice pest problems and respective
pesticide control measures reported by farmers, summer-autumn
crop, 1994, Long An, Vietnam.
Farmers’ responses Farmers who
Most important (n=603) applied
rice pest problems
pesticides
No. % for control (%)
Brown planthopper 164 27.2 97.0
Sheath blight 159 26.4 97.5
Leaffolder 99 16.4 94.9
Yellow leaf disease 60 10.0 98.3
Stem borer 45 7.5 100.0
Thrips 18 3.0 94.4
Cutworm 17 2.8 82.4
Caseworm 11 1.8 100.0
Rice bug 10 1.7 100.0
Blast 9 1.5 100.0
Grasshopper 5 0.8 60.0
Armyworm 3 0.5 100.0
Tungro 2 0.3 100.0
Rats 1 0.2 0.0
Pest management knowledge and perceptions
More than 80% of the farmers interviewed believed that pesticides would kill 75–
100% of the pests in their ricefields, but about 73% did not know that killing natural
enemies can induce population buildups of other pests such as the brown planthopper.
Leaf-feeding insects were the most frequently sprayed; the most common ones that
farmers mentioned were leaffolders (84%), caseworms (56%), armyworms (46%),
and thrips (23%). Most farmers perceived that leaf-feeding insects would cause severe
damage and yield losses to the rice crop (Fig. 3). They also believed that it was
necessary to spray early to control these pests. We compared rice yields reported by
farmers who sprayed early (during the first month of the summer-autumn crop
of 1994) and those who sprayed later (after the first month) using the Student t-test
and found them to be not significantly different (P = 0.3). The mean yields were
3.4 t ha -1 (n = 578, SD = 1.09) and 3.8 t ha -1 (n = 49, SD = 1.21), respectively.
Discussion
Farmers’ pest management practices in Long An did not appear to be different from
those of farmers interviewed in the Mekong Delta in May 1992 (Heong et al 1994, Vo
Mai et al 1993). The number of sprays farmer -1 season -1 dropped from 7 in 1992 to 5
in 1994. Proportionally more fungicides were used in 1994, increasing from 3% to
28%. Insecticide sprays dropped from 6 to 4 farmer -1 season -1 . But insecticide use
patterns were rather similar (Table 5). Organophosphates and carbamates were the
main insecticides used. Highly hazardous WHO category I compounds constituted
48% of the insecticide sprays in 1994, versus 37% in 1992. The main insecticides
used were methyl parathion (12%), methamidophos (29%), and monocrotophos (7%).
Pest management perceptions and practices in Long An Province, Vietnam 219

Fig. 3. Farmers’ perceptions of leaf-feeding insects (LF), Long
An, Vietnam.
Table 5. Comparison of pesticide sprays used in
the Mekong Delta and Long An.
Spray patterns
Mekong Delta, Long An,
1992 1994
Respondents (no.) 685 633
Sprays season -1 (no.) 4,704 3,033
Sprays farmer 1 season -1 7 5
(av no.)
Insecticide sprays (%) 89 70
Fungicide sprays (%) 3 28
Herbicide sprays (%) 8 2
Insecticide sprays 6 4
farmer 1 season -1
(av no.)
Type of insecticides
Organochlorines (%) 1 4
Organophosphates (%) 44 49
Carbamates (%) 32 20
Pyrethroids (%) 12 10
Others (%) 10 17
Classification by hazards
Category la (%) 17 12
Category Ib (%) 20 36
Category II (%) 59 38
Category V (%) 1 8
Unclassified (%) 3 6
220 Vo Mai et al

Table 6. Comparison of target pests of insecticide
sprays in the Mekong Delta (1992) and
Long An (1994).
Target pests
Sprays (%)
Mekong Long
Delta An
Rice leaffolders 20.5 33.4
Other lepidopterous 21.2 10.2
leaf feeders
Brown planthoppers 33.8 28.0
Stem borers 8.1 10.1
Thrips 4.1 14.0
Rice bugs 1.1 4.2
Other pests 14.2 0.1
Farmers’ target insect pests—mainly leaf-feeding insects and BPH—had remained
unchanged (Table 6). The main insecticides used to control leaf-feeding insects were
still methamidophos and methyl parathion, whereas carbamates were used to control
BPH. In addition, farmers’ first insecticide applications were mainly in the first 4 wk
after crop establishment, similar to that found in 1992.
From May 1992 to August 1994, the Ministry of Agriculture, together with FAO,
started implementing farmer field schools (FFS) in Vietnam. This farmer training
program has adopted the same model used in Indonesia (Matteson et al 1994). Farmers,
after the FFS training, were found to reduce their insecticide inputs significantly
(Matteson et al 1994). In Tan Tru District, the first FFS was conducted in the wet
season 1993 and, in August 1994, 11 FFS were conducted and 297 farmers trained. In
Tan Thanh District, the FFS started in late 1994 and, in August 1994, 6 FFS were
implemented and 170 farmers trained. The total population of rice farmers in these
two districts was about 20,000 and fewer than 3% of the farmers had undergone training
at the time of the survey. Therefore, this could not have accounted for the reduction
in insecticide use between 1992 and 1994. Moreover, farmers’ spray patterns had
not changed; farmers continued to apply insecticides early in the crop season to control
leaf-feeding insects.
Lepidopterous larvae, particularly rice leaffolders, continue to be farmers’ main
pest targets. Most farmers still perceive that leaf-feeding insects cause severe damage
and yield losses, and that early season spraying is necessary. But field infestations
with as much as 67% damaged leaves did not cause any yield loss (Miyashita 1985).
Using computer simulations, Fabellar et al (1994) found that leaf consumption by 20
larvae hill -1 was insufficient to significantly reduce crop yield. The average larval
densities in rice crops, on the other hand, are far below these densities. In the Philippines,
populations were under 1 hill -1 (Guo 1990) and, in Japan, the average was
below 2 (Wada and Shimazu 1978). Populations rarely reached 5 larvae hill -1 . Practically
no evidence shows that early season sprays against rice leaffolders are economi-
Pest management perceptions and practices in Long An Province, Vietnam 221

cally justifiable. Indeed, when farmers evaluated early season spraying, yields of their
plots with and without early insecticide sprays were not significantly different (Heong
et al 1995, Heong and Escalada 1998). Thus, the high proportion of insecticides used
by Long An farmers was still based on perceived rather than real needs.
The main insecticides farmers used had broad-spectrum mortality effects, which
have been shown to have detrimental effects on ecological balance (Cohen et al 1994,
Way and Heong 1994). These insecticides could cause secondary pest problems, such
as brown planthopper (Kenmore et al 1984, Heinrichs and Mochida 1984, Schoenly
et al 1994). In addition, the insecticides in WHO category I are hazardous to human
health (Rola and Pingali 1993).
Among diseases, farmers perceived sheath blight and yellow leaf disease as the
most important. Fungicides were used mainly when disease symptoms were visible.
For sheath blight, fungicides might have worked, because the chemicals used—
hexaconazole and validamycin—can control the disease. For effective control, sprays
had to be timed appropriately. The survey, however, did not solicit fungicide timing
information. The causal factor of yellow leaf disease is still controversial. Farmers
had perceived the symptom to be a disease and had adopted the use of fungicides. But
whether the fungicide sprays used had prevented yield loss is uncertain. Clearly, scientists
need to better understand farmers’ knowledge, attitudes, and practices in managing
diseases in order to improve decision making. As pointed out by Bentley (1992),
because diseases are important but difficult to observe, scientists have opportunities
to help expand existing folk taxonomies, enhance farmer observations, and challenge
existing beliefs.
The insect pest management practices of Long An farmers were found to be far
from satisfactory because unnecessary insecticide use was widespread. An obvious
need is farmer training. At the time of the survey, only 6% of the farmers interviewed
said they had attended training. If higher proportions of farmers were trained, unnecessary
spraying could decrease. Farmer training through the FFS approach has been
found to be effective (Matteson et al 1994). In Indonesia, where this training method
was introduced in 1989, 200,000 farmers (or 1.3%) were trained in 4 yr (van de Fliert
et al 1995). If FFS training were to cover the estimated 15 million farmers in Indonesia
and the current training rate were maintained, it would take 18 yr to train all 15
million.
In August 1994, 83 FFS began in Long An and 2,000 farmers were trained. By
the end of 1996, 8,000 (or 4% of the province’s estimated 200,000 farmer families)
were expected to be trained. To reach the remaining 96%, a drastic increase in the
number of FFS being implemented and financing would be needed. Meanwhile, other
options with a wider and faster reach need to be evaluated. Unnecessary insecticide
applications brought about primarily by misperceptions will continue to plague farmers
unless alternative information that can change their perceptions reaches them. The
use of strategic extension campaigns effectively influenced farmers’ rat control and
weed control practices in Malaysia (Adhikarya 1994, Matteson et al 1994). Although
an initial 5% increase in herbicide use was noted immediately after the weed management
campaign in 1989 (Matteson et al 1994), herbicide use dropped by 33% in 1993
222 Vo Mai et al

(Adhikarya 1994). In addition, farmers’ land leveling practices contributed to better
weed management. The cost-benefit ratio of the integrated weed management campaign
was estimated at 1:50. The use of media perhaps needs to be re-examined.
These methods of communication use systems readily available to farmers and can
disseminate information rapidly.
The problem at hand seems to be more related to misperceptions and attitudes of
farmers. Research on the content, format, and mechanisms to communicate different
information on pest management in various sociocultural and political contexts may
be useful.
Responses to media across cultures may vary because of differences in visual
literacy. For instance, in Honduras, farmers can be distracted and sometimes offended
by comic-book pamphlets (Goodell et al 1990). Farmers preferred pamphlets with
naturalistic line drawings and clear technical prose (Bentley and Andrews 1991). Discovering
the critical misperceptions and attitudes and identifying mechanisms for
effecting a rapid change in pest management practices are still important and immediate
challenges.
References
Adhikarya R. 1994. Strategic extension campaign: a participatory-oriented method of agricultural
extension. Rome: FAO.
Bentley JW. 1992. Alternatives to pesticides in Central America: applied studies of local knowledge.
Cult. Agric. 14:10-13.
Bentley JW, Andrews K. 1991. Pests, peasants and publications: anthropological and entomological
views of an integrated pest management program for small-scale Honduran farmers.
Human Org. 50:113-124.
CIRAD (International Cooperation Centre in Agronomic Research for Development). 1991.
Regional agro-pesticide index, Volume 1 : Asia. 3rd ed. Paris (France): CIRAD.
Cohen JE, Schoenly K, Heong KL, Justo H, Ariga G, Barrion AT, Litsinger JA. 1994. A food
web approach to evaluating the effect of insecticide spraying on insect pest population
dynamics in a Philippine irrigated rice ecosystem. J. Appl. Ecol. 31:747-763.
Conway GR, Pretty JN. 1991. Unwelcome harvest: agriculture and pollution. London: Earthscan
Publications.
Cue NS. 1995. Agriculture of Vietnam 1945-199s. Hanoi (Vietnam): Statistical Publishing
House. 386 p.
Escalada MM, Heong JSL. 1993. Communication and implementation of change in crop protection.
In: Crop protection and sustainable agriculture. CIBA Foundation Symposium
177. Chichester: J. Wiley & Sons. p 191-207.
Fabellar LT, Fabellar N, Heong KL. 1994. Simulating rice leaffolder feeding effects on yield
using MACROS. Int. Rice Res. Newsl. 19(2):7-8.
Goodell GE, Andrews KL, Lopez J. 1990. The contribution of agronomo-anthropologist to onfarm
research and extension in integrated pest management. Agric. Sys. J. 33:321-340.
Guo Y. 1990. Larval parasitization of rice leaffolders (Lepidoptera: Pyralidae) under field and
laboratory conditions. PhD thesis. University of the Philippines, Los Baños.
Heinrichs EA, Mochida, O. 1984. From secondary to major pest status: the case of insecticideinduced
rice brown planthopper, Nilaparvata lugens, resurgence. Protect. Ecol. 7:201-
218.
Pest management perceptions and practices in Long An Province, Vietnam 223

Heong KL, Escalada MM. 1998. Changing farmers’ pest management practices through farmer
experimentation. Int. J. Pest Manage. (In press.)
Heong KL, Cuc NTT, Binh N, Chien HV, Fujisaka S, Bottrell DG. 1995. Reducing early season
insecticide applications through farmers’ experiments in Vietnam. In: Denning GL,
Xuan VT, editors. Vietnam and IRRI: a partnership in rice research. Manila (Philippines):
International Rice Research Institute, and Hanoi (Vietnam): Ministry of Agriculture and
Food Industry. p 217-222.
Heong KL, Escalada MM, Mai V. 1994. An analysis of insecticide use on rice: case studies in
the Philippines and Vietnam. Int. J. Pest Manage. 40(2):173-178.
Kenmore PE, Cariño FO, Perez CA, Dyck VA, Gutierrez AP. 1984. Population regulation of the
rice brown planthopper (Nilaparvata lugens Stal.) within rice fields in the Philippines. J.
Plant Protect. Trop. 1:19-38.
Mai V, Cuc NTT, Hung NQ, Chau LM, Chien HV, Escalada MM, Heong KL. 1993. Farmers’
perceptions of rice pest problems and management tactics used in Vietnam. Int. Rice Res.
Newsl. 18(2):31.
Matteson PC, Gallagher KD, Kenmore, PE. 1994. Extension of integrated pest management
for planthoppers in Asian irrigated rice: empowering the user. In: Denno R, Perfect TJ,
editors. Planthoppers: their ecology and management. London: Chapman and Hall. p 656-
685.
Miyashita T. 1985. Estimation of the economic injury level in the rice leaffolder, Cnaphalocrocis
medinalis Guenee (Lepidoptera: Pyralidae). I. Relation between yield loss and injury of
rice leaves at heading or in the grain filling period. Jpn. J. Appl. Entomol. Zool. 29:73-76.
Pingali PL, Xuan VT. 1992. Vietnam: decollectivization and rice productivity growth. Econ.
Devel. Cult. Change 40:697-718.
Rola A, Pingali PL. 1993. Pesticides, rice productivity and farmers’ health: an economic assessment.
Washington (D.C.): World Resources Institute, and Manila (Philippines): International
Rice Research Institute.
Schoenly KG, Cohen JE, Heong KL, Arida GS, Bamon AT, Litsinger JA. 1994. Quantifying
the impact of insecticides on food web structure of rice-arthropod populations in a Philippine
farmer’s irrigated field: a case study. In: Polis GA. Wisemiller K, editors. Food webs:
integration of patterns and dynamics. London: Chapman and Hall. p 343-351.
SAS (Statistical Analysis System). 1985. SAS user’s guide statistics. Version 5 edition. Cary
(N.C., USA): SAS Institute, Inc.
van de Fliert E, Pontius J, Roling N. 1995. Searching for strategies to replicate a successful
extension approach: training of IPM trainees in Indonesia. Eur. J. Agric. Educ. Ext. 1:41-
63.
Wada T, Shimazu M. 1978. Seasonal population trends of the rice leaf roller, Cnaphalocrocis
medinalis Guenee (Lepidoptera: Pyralidae). Appl. Entomol. Zool. 17:278-281.
Way MJ, Heong KL. 1994. The role of biodiversity in the dynamics and management of insect
pests of tropical irrigated rice—a review. Bull. Entomol. Res. 84:567-587.
224 Vo Mai et al

Notes
Authors’ addresses: Vo Mai and N.H. Huan. Department of Plant Protection, Ministry of Agriculture
and Rural Development, S.R. Vietnam; K.L. Heong and A.A. Lazaro. Entomology
and Plant Pathology Division, International Rice Research Institute, P.O. Box 933, 1099
Manila, Philippines; M.M. Escalada, Department of Development Communication, Visayas
State College of Agriculture, Baybay, Leyte, Philippines.
Acknowledgments: The authors wish to express their appreciation to the Swiss Agency for
Development and Cooperation, which supported this research through the Rice Integrated
Pest Management Network, coordinated by the International Rice Research Institute in
Los Baños, Philippines. Many people contributed to this research. In particular, the authors
are grateful to Madam Sui, the Vice Chairperson of the People's Committee of Long
An Province, and Madam Ngoc and her staff. who helped implement the survey.
Citation: Heong KL, Escalada MM, editors. 1997. Pest management of rice farmers in Asia.
Manila (Philippines): International Rice Research Institute.
Pest management perceptions and practices in Long An Province, Vietnam 225

CHAPTER 16
A comparative analysis
of pest management practices
of rice farmers in Asia
K.L. Heong and M.M. Escalada
Farmer surveys in 10 Asian countries were used to examine trends
in rice pest management practices. Pesticides remain the dominant
control tactic of farmers, who use insecticides more frequently than
herbicides and fungicides. A large proportion of farmers were still
using insecticide compounds classified as extremely or highly hazardous
to human health (WHO I)—namely, methyl parathion,
monocrotophos, and methamidophos. Herbicides were commonly
used in Indonesia, Malaysia, Sri Lanka, China, Vietnam, and the
Philippines. Only farmers in China and Vietnam used fungicides regularly.
The mean number of insecticide sprays varied from 0.3 in Laos
to 3.9 in Vietnam. Most of these sprays were applied during the
seedling, tillering, and booting stages of the rice crop. Farmers usually
overreacted to leaf-feeding pests, collectively referred to as
“worms,” and tended to apply their first insecticide sprays during
the first 4 weeks after crop establishment. They strongly believed
that leaf-feeding insect pests were damaging and reduced yield. Based
on this perception, farmers chose Insecticides (or medicine) to kill
pests to protect their yields. To improve farmers’ pest management
perceptions and decision making, research needs to address issues
such as the influence of communication media on perceptions and
attitudes.
Introduction
The Green Revolution in Asia during the 1960s and 1970s had a dramatic impact on
the rice economy. Despite limits to land availability, annual growth in rice production
in Asia was about 2.7% from 1966 to 1993. Production doubled from 240 million t in
1966 to 482 million t in 1993 (IRRI 1995). The major source of growth had been from
yield ha -1 , which increased from 2.0 to 3.7 t ha -1 over the same period. In some countries,
yields doubled, from 1.8 to 4.4 t ha -1 in Indonesia, from 1.3 to 2.8 t ha -1 in the
Philippines, from 1.3 to 2.7 t ha -1 in India, and from 1.8 to 3.5 t ha -1 in Vietnam. These
increases had been attributed to the widespread adoption of modern high-yielding
221

varieties (HYVs) introduced in 1966. In South and Southeast Asia, the adoption rate
of HYVs reached 60% by the end of the 1980s (David and Otsuka 1994).
The rapid spread of the modern varieties had been facilitated by national production
programs that disseminated the new technology and provided irrigation facilities
and credit to farmers. In the Philippines, the Rice Sufficiency Program and the
Masagana 99 Program began in 1966 and 1973, respectively (Barker et al 1978). A
similar program, BIMAS, began in Indonesia (Adjid 1983). In both countries, the
government introduced fertilizer and pesticide credits together with the HYVs. Routine
fertilizer and insecticide applications were deemed necessary inputs and, as a
result, farmers sprayed insecticides as many as 15 times a season. Although IR8, the
first HYV introduced, was susceptible to pests, subsequent varieties incorporated
multiple insect and disease resistance (Panda and Khush 1995). But farmers were
encouraged to use insecticides at regular intervals, even when growing these varieties,
through subsidies and loan schemes (Kenmore et al 1987, Conway and Barbier
1990, Conway and Pretty 1991).
High yields from insecticide-protected fields at experiment stations (Pathak and
Khan 1994) and high crop losses attributed to insects (Cramer 1967, IRRI 1990)
encouraged the chemical industry to invest in new insecticide development programs
and influenced adoption through advertising and marketing strategies. In the Philippines,
for instance, most farmers used insecticides despite adopting new HYVs with
multiple pest resistance (IRRI 1994). Research, on the other hand, focused on technology
generation rather than on solving farmers’ problems. Limited efforts went into
understanding farmers’ knowledge, perceptions, attitudes, and practices in pest control.
Recent analyses of farmers’ pest management in the Philippines and Vietnam
(Heong et al 1994, 1995a,b) showed that a large proportion of insecticide sprays that
farmers used did not produce economic returns. Instead, they could be detrimental to
the ecological balance and cause secondary pest problems, such as the brown
planthopper (Cohen et al 1994, Schoenly et al 1994, Way and Heong 1994). In addition,
pesticides pose threats to farmers’ health (Rola and Pingali 1993). Between 1991
and 1996, farmer surveys to document farmers’ pest management knowledge, attitudes,
and practices were carried out in 10 Asian countries. In this paper, we have
extracted a few indicator variables of these practices from the surveys and made a
comparative analysis.
Each farmer survey was conceptualized and conducted independently by the respective
researchers; in some cases, the generic survey questionnaire developed by
the Rice IPM Network for an earlier survey in Leyte, Philippines (Heong et al 1995a),
served as a guide. Except for the survey in Thailand. which focused on women’s
perceptions and practices, nearly all of the farmer surveys in the analysis—Cambodia,
China, India, Indonesia, Laos, Malaysia, Philippines, and Sri Lanka—addressed
farmers’ pest management knowledge, attitudes, and practices. The surveys were conducted
in predominantly lowland rice areas, except in Cambodia and Laos, and covered
sample sizes ranging from 120 to 1,145 (Table 1). The average farm sizes of
228 Heong and Escalada

Table 1. Details of farmer surveys used in the analysis.
Sample Average Farmers’
Country Location Survey date size farm size mean age
(ha) (yr)
Cambodia Lowland rice area Jun 1995-Apr 1996 1,145 0.6 42.6
China Hunan Apr-May 1991 266 0.25 n.a.
India Tamil Nadu 1994-95 120 n.a. middle-aged
Indonesia West Java 1991-93 128 1.2 44.6
Laos Northern, central, Feb-Oct 1994 463 1.5 mostly 30-45
and southern regions
Malaysia Kedah and Kelantan 1991 and 1994 450 1.5 >40
Philippines Nueva Ecija June 1994 300

Table 3. Types of insecticides used by rice farmers in Asian countries.
Farmers (%) reporting use of these insecticides
Chemical family/
active ingredient WHO Camb China lndo Laos Malay Phil SLanka Thai Viet
category a
Organophosphates
Methyl parathion la 5.5 37.4 27.9 21.0 12.1
Metharnidophos Ib 39.8 2.8 0.8 0.5 1.5 0.8 28.9
Azinphos-ethyl Ib 3.2
Monocrotophos Ib 61.8 12.1 53.2 3.0 0.8 6.8
Fenthion lb 1.3 1.5
Phenthoate II 8.0
Chlorpyrifos II 13.8 3.7 7.0
Diazinon II 11.2 0.5 3.2 1.6
Profenofos - 0.5
Malathion III 1.8 0.9
Organochlorines
Endosulfan
Endrin
Lindane
II
la
II
2.8 7.8 49.5 2.5 8.8 4.0
0.5
0.1
Pyrethroids
Cypermethrin II
Deltamethrin II
Cyhalothrin –
Fenvalerate II
Esfenvalerate –
Carbamates
MTMC
II
Carbofuran Ib
Carbaryl
II
BPMC (fenobucarb) II
MlPC (isoprocarb) II
Carbosulfan II
Thiocarbamate –
Methomyl
Ib
2.0 13.1 52.7 0.8 2.0
18.8 14.7 0.8 3.8
9.6 2.9
0.1
0.1
22.2
38.1 12.7 4.6 11.8 21.9 0.1
0.5 6.1
42.0 12.9 0.9 2.7 16.5
70.3 2.6 0.4 2.7
1.2
6.1
9.6
Others
Cartap II 2.6
Buprofezin V 23.1 0.8 0.5 7.8
MTMC + fenthoate II 1.1
Chlordimeform II few
Azinphos + ethyl +
fenobucarb 1b 0.5
Chlorpyrifos +
fenobucarb II 4.6
Fenobucarb +
methyl parathion – 0.1
Not specified – 0.2
a la = extremely hazardous, Ib = highly hazardous, II = moderately hazardous, IIl = slightly hazardous, V = unlikely
to present acute hazard in normal use, – = unclassified.
Source: CIRAD 1991.
230 Heong and Escalada

Methyl parathion was used by farmers in Laos (37.4%), the Philippines (27.9%),
Thailand (21 %), and Vietnam (12.1 %). For methamidophos, most users were farmers
in China (39.8%) and Vietnam (28.9%), whereas monocrotophos was applied by more
farmers in Indonesia (61.8%), the Philippines (53.2%), and Cambodia (10.4%). Among
the organochlorines, endosulfan was the most popular, with the highest use recorded
for farmers in the Philippines (49.5%), followed by Thailand (8.8%), Malaysia (7.8%).
Vietnam (4%), Indonesia (2.8%), and Sri Lanka (2.5%). Cypermethrin and deltamethrin
were the most common pyrethroids used. The majority of cypermethrin users were
farmers in the Philippines (52.7%), followed by those in Malaysia (13.1%) and a few
in Indonesia (2%) and Vietnam (2%). Deltamethrin was used mainly in Indonesia
(18.8%) and the Philippines (14.7%).
Compared with other carbamates, carbofuran appeared to be popular among farmers
in Indonesia (38.1 %), Thailand (21.9%), Malaysia (12.7%), and Sri Lanka (11.8%),
followed by fenobucarb (BPMC), used mainly in Indonesia (42%) and Vietnam
(16.5%). MTMC and MIPC were used mostly by farmers in China.
Fungicides. The surveys revealed that only farmers interviewed in China and
Vietnam were heavy fungicide users. In China, 100% of the farmers reported using
Jinggang-mycin, a local fungicide. Vietnamese farmers indicated that they used a
variety of fungicides, such as hexaconazole (37.9%), validamycin (23.1 %), benomyl
(15.6%), carbendazim (4.9%), and propiconazole (4.6%) (see Vo Mai et al, this volume,
Chapter 15). In India, farmers were reported using edifenphos, but the exact
percentages of farmers using pesticides were not specified.
Herbicides. Based on the data available, farmers in Malaysia and Indonesia were
high users of herbicides (Table 4). Although farmers in Cambodia and Sri Lanka
reported using them, the specific compounds used were not indicated. For herbicides,
Indonesian farmers preferred metsulfuron-methyl (75.1%) and 2,4-D (38.1 %), whereas
farmers in Malaysia preferred 2,4-D (46.4%) and molinate + 2,4-D (20.4%). In the
Philippines, pretilachlor and piperophos + 2,4-D were the most common herbicides.
Rodenticides. In the 10 countries, more farmers in Indonesia, Malaysia, and Cambodia
used rodenticides. Zinc phosphide was most commonly used by farmers in
Cambodia and Malaysia, whereas brodifacoum was mainly used in Malaysia. To control
rats in irrigated and tidal swamp areas, 87.3% of the farmers in Indonesia used the
insecticide aldicarb.
Farmers used some pesticides that were extremely or highly hazardous to human
health. Table 5 shows the proportions of farmers interviewed who used pesticides in
the various WHO classifications. In Laos, the Philippines, and Thailand, many farmers
were still using methyl parathion, classified as WHO Ia (extremely hazardous).
Highly hazardous WHO Ib compounds were reportedly used by more fanners in
Malaysia, the Philippines, China. Vietnam, Thailand and Sri Lanka. These pesticides
were mainly methamidophos, monocrotophos, and carbofuran. About 60% of the farmers
in Malaysia reported using zinc phosphide, a WHO Ib compound, in rat baits.
A comparative analysis of pest management practices of rice farmers in Asia 231

Table 4. Types of herbicides and rodenticides used by rice farmers in
Asian countries.
Chemical family/
active ingredient
Farmers (%) reporting use of these pesticides
WHO a Camb lndon Malay Phil b SLanka
Herbicides
Bensulfuron-methyl – 0.6 0.2 2.5
Molinate + 2,4-D II 20.4
Quinclorac – 0.6
Propanil Ill 1.8
Pretilachlor – 1.3 34.7
2,4-D II 38.1 46.4 6.5
Metsulfuron-methyl V 75.1 8.4
Oxadiazon V 5.5
Butachlor V 8.1
Butachlor + propanil 7.2
Piperophos + 2,4-D 28.7
Thiobencarb + 2,4-D 0.5
MCPA III 17.7 6.2
Fexanoprop – 7.1
Cinosulfuron – 1.7
Not specified – 1.7
Rodenticides
Aldicarb c la 87.3
Zinc phosphide Ib 8.6 2.2 43.8
Brodifacoum la 24.7 0.7
Bromidialone la 2.7
Chlorophacinone la 5.6
Coumatetralyl Ib 1.1
Warfarin Ib 0.2
a la = extremely hazardous, Ib = highly hazardous, II = moderately hazardous, III
= slightly hazardous, V = unlikely to present acute hazard in normal use, – =
unclassified. b Figures based on herbicide applications (%) from Moody et al (this
volume, Chapter 10). c This insecticide, called Temik 106, is widely used by
farmers in Indonesia as a rat poison.
Source: CIRAD 1991.
Table 5. Classification of pesticides used by rice farmers in Asian countries.
Farmers (%) using pesticides
WHO classification
by hazard Camb China Laos Malay Phil SLanka Thai Viet
Extremely hazardous (la) 9.5 – a 8.6 28.3 b 28.4 0.7 21.0 12.0
Highly hazardous (Ib) 1.7 39.8 5.8 59.9 c 56.4 22.2 23.5 36.0
Moderately hazardous (II)
– 70.3 9.0 76.1 83.5 29.7 16.5 38.0
Slightly hazardous (Ill)
– – – 1.8 1.8 0.9 – –
Unlikely to present acute – – – 10.5 – – – 8.0
hazard in normal use (V)
Unclassified (–) – 54.5 – – 15.6 – – 6.0
a – = negligible.
b Mainly brodifacoum as rat poison.
c
Mainly zinc phosphide as rat poison and
carbofuran.
232 Heong and Escalada

Pests reported to be most important by farmers
Insect pests dominated the list of pests reported by rice farmers in the 10 countries
surveyed, followed by weeds, rats, and diseases. Table 6 lists the pests mentioned as
important by farmers. In 9 out of the 10 countries—Cambodia, China, India, Laos,
Malaysia, Philippines, Sri Lanka, Thailand, and Vietnam—rice farmers reported lepidopterous
leaf feeders to be the most important pests, followed by the stem borer,
brown planthopper, weeds, rice bug, rats, and birds.
Farmers’ pesticide spray targets
Farmers’ pesticide applications in Asia mainly targeted insect pests and weeds (Table
7). Among the insects, lepidopterous leaf feeders, stem borers, brown planthopper,
and rice bugs appeared to be commonly sprayed. Rice farmers in China, Indonesia,
Malaysia, Sri Lanka, Thailand, and Vietnam also commonly used herbicides. For disease
problems, fungicides were often used only in China and Vietnam. Their main
targets were sheath blight, blast, and yellow leaf disease.
Frequency and timing of insecticide applications
Insecticides were applied as many as 11 times a season in some countries (e.g., Vietnam),
whereas in Cambodia and Laos, many farmers (63% and 77%, respectively)
did not apply any. According to the frequency distributions of applications in some
countries (Table 8), Vietnamese farmers applied an average of 4 sprays per season,
whereas in Laos the average was only 0.3. We used both the mean and the mode to
describe the central tendency of the farmers’ number of insecticide sprays. There was
Table 6. Pests reported to be important by farmers in different Asian countries.
Pests
Countries
Camb China India Indo Laos Malay Phil SLanka Thai Viet
Insects
Lepidopterous
leaf feeders
Stem borers
Rice bugs
Brown planthopper
Thrips
Gall midge
Diseases
Sheath blight
Blast
Bacterial leaf blight
Yellow leaf disease a
Tungro
Weeds
Rats
Birds
a Casual agent still unknown.
A comparative analysis of pest management practices of rice farmers in Asia 233

Table 7. Main target pests for which pesticides were used by rice farmers in Asia.
Pests
Insects
Lepidopterous
leaf feeders
Stem borers
Rice bugs
Black bug
Brown planthopper
Thrips
Diseases
Sheath blight
Blast
Yellow leaf disease
Tungro
Weeds
Rats
Birds
Snails
Farmers (%) reporting these target pests
Camb China India lndo Laos Malay Phil SLanka Thai
Viet
Table 8. Frequency of insecticide use of rice farmers in Asian countries.
Sprays (no.)
Camb
Farmers (%) reporting
lndo Laos Malay Phil SLanka Thai Viet
0
1
2
3
63.1
14.2
11.7
7.5
3.9
14.9
27.6
19.9
76.9
18.2
4.1
0.6
4.8
25.9
47.0
20.6
sprays season -1
4
5
6
7
8
9
10
11
Mean no. of
Mode
Variance (S 2 )
2.8
0
0
0.4
0
0
0
0
0.7
0
137.8
25.4
7.2
1.1
0
0
0
0
0
2.7
2
179.6
0
0
0
0
0
0
0.2
0
0.3
0
31.5
1.7
0
0
0
0
0
0
0
1.9
2
71.2
5.3
4.4
18.9
35.5
14.9
9.6
7.9
1.7
0.4
0.9
0.4
0
3.3
3
248.0
36.4
21.5
18.8
11.8
11.5
0
0
0
0
0
0
0
1.4
0
189.5
0
35.4
31.8
21.9
4.7
1.0
2.6
1.0
0
0
1.6
0
2.3
1
86.8
1.1
4.4
13.6
22.5
28.2
16.5
8.5
3.0
1.6
0.6
0
0.2
3.9
4
232.9
234 Heong and Escalada

Table 9. Timing of insecticide applications of rice farmers in Asian countries.
Crop stage
Seedbed
Seedling
Tillering
Booting
Flowering
Milking
Soft dough
Ripening
Farmers (%) a
Camb lndon Laos Malay Phil SLanka Thai Viet
0 0 2.4 0 0 0 0 0
5.6 30.3 17.0 38.7 26.6 42.0 44.7 57.8
27.9 75.9 0 57.3 41.5 40.0 28.7 83.4
14.0 33.4 2.4 6.7 17.5 18.0 12.7 86.2
7.5 13.8 0 33.7 4.4 0 7.7 67.7
7.5 13.8 0.9 21.9 3.4 0 2.4 25.6
0 0.6 0 0 1.1 0 3.0 0.6
0 0 0 0 0 0 0.9 0.2
a Farmers sprayed at more than one crop stage.
Table 10. Timing of first insecticide applications of rice farmers
in Asian countries.
Wk after crop
establishment
Farmers (%)
Indonesia Malaysia Philippines Vietnam
1–4 59.7 53.9 97.8 84.1
5–8 29.8 23.5 2.2 15.1
>8 1.7 1.1 0.0 0.8
a high variation among farmers in most countries, with the Philippines and Vietnam
having the highest. In Laos, where variance was low, most farmers did not apply any
insecticides. In Malaysia, 47% applied two sprays and in Thailand 35% applied one.
Most pesticide applications were made in the early stages of the crop (Table 9).
Except for Cambodia and Laos, some of these were herbicides. At the tillering stage,
most farmers who sprayed their crops were in Vietnam (83.4%), Indonesia (75.98),
and Malaysia (57.3%). At the booting stage, most farmers who applied pesticides
were from Vietnam (86.2%). In most of these cases, the main pesticides used were
insecticides.
Data were then disaggregated to examine the distribution of farmers’ first insecticide
applications. Among the four countries where data were available (Table 10),
many farmers applied their first sprays during the first 4 wk after crop establishment,
especially in the Philippines (97.8%) and Vietnam (84.1%).
Most of these early crop insecticide sprays were for leaf-feeding insects. When
examining farmers’ perceptions of leaf-feeding insects, we found an interesting association
between perceptions and behavior. Table 11 shows that a large proportion of
farmers in Sri Lanka (72%), Vietnam (66.7%), Laos (66%), and the Philippines (60%)
believe that leaf-feeding insects cause severe damage to the rice crop. Similarly, a
majority of the farmers in six countries—China (93.6%), Laos (72.5%), Vietnam
(69.7%), Sri Lanka (65.2%), the Philippines (52.3%), and Thailand (51.4%)—believe
that leaf-feeding insects cause yield losses and that it is necessary to control
them early in the cropping season.
A comparative analysis of pest management practices of rice farmers in Asia 235

Table 11. Farmers’ perceptions of leaf-feeding insects in Asian countries.
Attitude statement China Laos Philippines Sri Lanka Thailand Vietnam
Leaf-feeding insects n.a. a 66.0 60.0 72.0 41.2 66.7
cause severe damage
to the rice crop.
Leaf-feeding insects 93.6 72.5 52.3 65.2 51.4 69.7
cause yield loss
It is necessary to 93.6 n.a. 97.6 60.3 43.4 n.a.
spray for leaf-feeding
insects.
Control for leaf-feeding 93.6 43.5 57.6 56.5 36.2 76.9
insects should be
done early.
a n.a. = not available.
Discussion
From a total of 4,055 farmers surveyed and the hundreds that we interviewed personally,
several trends in Asian rice farmers’ pest management are evident. Pesticides
remain the dominant control tactic that farmers rely on. This might be attributed partly
to their association of pesticides with medicine. Indeed, pesticides are referred to as
medicine in all the countries we surveyed (Appendix I). The surveys showed that rice
farmers’ pest management practices have changed little since 1984 from the description
of Lim and Heong (1984). In some cases, such as Malaysia, the proportion of
farmers using insecticides rose from 46% in 1975 to 62% in 1984 (Heong et al 1985)
and 95% in 1995 (Nonniyah and Chang, this volume, Chapter 8). The chemicals
changed, from less use of endosulfan to increased use of cypermethrin. In other cases,
however, such as the Philippines, farmers were still using chemicals they had used 15
years ago (Litsinger et al 1980).
Pesticides used by farmers
Farmers used insecticides more frequently than any other pesticides, followed by
herbicides and fungicides. This distribution of pesticide use is reflected in the rice
pesticide market outside Japan for the period 1990-95 (Fig. 1) (McKenzie 1996). The
total market value was US$1.6 billion. Insecticides accounted for 47%, herbicides
31%, and fungicides 22%. Large proportions of farmers surveyed in the Philippines,
China, Thailand, and Vietnam were still using pesticides classified by WHO as extremely
hazardous (WHO Ia) or highly hazardous (WHO Ib) to human health, banned,
or under restricted use in many countries.
Many farmers were still using insecticide compounds banned or restricted because
of their potential hazards to human health. Methyl parathion (WHO Ia),
monocrotophos (WHO Ib), and methamidophos (WHO Ib) were commonly used in
Cambodia, China, India, Indonesia, Laos, the Philippines, Thailand, and Vietnam.
Although banned by governments, these products remain readily available to farmers
in Indonesia, the Philippines, and Vietnam, whereas governments of the other coun-
236 Heong and Escalada

Fig. 1. The rice pesticide market outside Japan, 1990-95 (data from
McKenzie Consultants Ltd.).
tries had not categorically banned them. In Malaysia, rice farmers hardly used any of
these products, perhaps because of strong government commitments to implement
the Pesticides Act.
Pyrethroids, particularly cypermethrin and deltamethrin, were commonly applied
to Asian ricefields. These chemicals mainly targeted leaf-feeding insects, such as rice
leaffolders. Though frequently sprayed, however, these pests cause negligible yield
losses (Heong 1993). Damage by these pests can be highly visible, especially during
the early crop stages, which stimulates farmers to respond by spraying. These early
insecticide applications may, however, favor the development of secondary pests, such
as the brown planthopper (BPH), Nilaparvata lugens. Evidence that BPH is a secondary
pest of rice is well documented (Kenmore et al 1984, Heinrichs and Mochida
1984, Schoenly et al 1994, Rombach and Gallagher 1994, Way and Heong 1994) and
pyrethroid insecticides have been shown to disorganize ecological food web relationships
that favor planthoppers (Cohen et al 1994).
Endosulfan was the main organochlorine often used by Asian farmers. Although
endosulfan was banned in the Philippines, about 50% of the farmers surveyed said
they were still using it. Endosulfan was used to control the golden apple snail, besides
controlling insect pests. Farmers in China, Indonesia, Malaysia, the Philippines, and
Vietnam frequently used carbamates. The most common chemicals were BPMC and
MIPC, used mainly to control planthoppers and leafhoppers. Another insecticide used,
mainly in Indonesia, for hopper control was buprofezin. In Indonesia, Thailand, Sri
Lanka, and Malaysia, farmers used carbofuran granules for stem borer control. Another
carbamate, aldicarb, categorized as an extremely hazardous pesticide (WHO
Ia), was a common ingredient in rat poisons in Indonesia.
Farmers commonly used herbicides in Cambodia, Indonesia, Malaysia, Sri Lanka,
China, Vietnam, and the Philippines. According to results from Indonesia, Malaysia,
and the Philippines (Moody et al, this volume, Chapter 10), farmers mainly used the
A comparative analysis of pest management practices of rice farmers in Asia 237

herbicides 2,4-D, pretilachlor, metsulfuron-methyl, piperophos + 2,4-D, and MCPA.
In the Philippines, most farmers made one or two applications, mainly during the
seedling stage, to control Echinochloa species and other weeds.
Of the 10 countries surveyed, only farmers in China and Vietnam used fungicides
regularly. In China, a locally manufactured antibiotic, Jinggang-mycin, was commonly
applied to ricefields, mainly to control sheath blight. Similarly, many farmers in Vietnam
used validamycin to manage sheath blight. Two other fungicides frequently used
in Vietnam were hexaconazole and benomyl to control sheath blight and yellow leaf
disease (Vo Mai et al, this volume, Chapter 15). Although the causal agent of yellow
leaf disease had not been determined, farmers were responding to the symptoms by
spraying fungicides. Because about 98% of the farmers interviewed in Long An had
applied fungicides to control this disease, this is an opportunity for research to help
farmers avoid pesticide misuse.
The mean number of insecticide sprays varied from 0.3 in Laos to 3.9 in Vietnam.
In Cambodia and Laos, more than 60% of the farmers did not use any insecticides
and in Sri Lanka 36% did not apply any. In the other countries, farmers usually
applied 2 or 3 sprays. Most of these sprays were applied during the seedling, tillering,
and booting stages of the rice crop. At these stages, the main pests that are highly
visible are leaf-feeding insects, such as rice leaffolders, the rice caseworm, thrips,
whorl maggots, and armyworms. Under extreme conditions, each of these species
can destroy a rice crop, but such conditions in nature are extremely rare. The whorl
maggot usually attacks rice in the first 2 wk after transplanting. Although as much as
100% of the hills can be damaged, causing the crop to look devastated, no yield loss
could be detected (Viajante and Heinrichs 1986). Similarly, for leaffolders, normal
larval densities are under 5 hill -1 ; it was shown that densities need to reach 15 before
any yield loss can be detected (Fabellar et al 1994).
Farmers’ perceptions of pests
In the 10 countries surveyed, farmers’ overreactions to pests—collectively referred to
as “worms”—were rather similar to those of an earlier analysis of farmers’ pest management
practices in the Mekong Delta of Vietnam and Leyte Province in the Philippines
(Heong et al 1994). In countries where data were available, we found that farmers
tended to apply their first insecticides, usually to control leaf-feeding insects, during
the first 4 wk after crop establishment. Farmers also strongly believed that leaffeeding
insects would cause yield losses. Perhaps Asian rice farmers in general have
similar attitudes and behavior toward pest management. They strongly believe that
leaf-feeding pests are damaging and yield-reducing. Based on this perception, they
choose insecticides (or medicine) to kill pests to protect their yields. Farmers seem to
emphasize “killing” pests rather than preventing losses. Over the years, farmers have
learned to use insecticides that are less expensive and highly toxic to “worms,” which
might account for the high use of methyl parathion and monocrotophos. Asian rice
farmers seem to be trapped in a vicious circle of making wrong decisions about pest
problems and insecticide use and have become victims of insecticide abuse (Bentley
1989). It is evident that, to improve farmers’ pest management decisions, we need to
238 Heong and Escalada

identify processes that can help change farmers’ perceptions of pests and their management.
From season to season, farmers are likely to face similar pest problems. Decisions
made in the previous season (period t) can influence the current season’s decisions
(period t+l) (Mumford and Norton 1984). Because pest problems, control options,
and objectives often remain much the same year after year, farmers’ perceptions
and actions rarely change. This could account for the persistence in farmers’
insecticide misuse. Since the outcome of period t can influence perceptions in the
next period, introducing an action that can result in a different outcome can be used to
change farmers’ perceptions. Farmers can also be persuaded to use other objectives,
such as yields and profits, rather than insect deaths, when comparing outcomes. This
approach was applied to change farmers’ perceptions and practices in rice leaffolder
control in the Philippines (Heong and Escalada 1997) and Vietnam (Heong et al 1995b).
Participating farmers reduced their insecticide use by 50% after conducting an experiment
to evaluate whether the heuristic (or simple rule)—spraying insecticides in
the first 30 d after transplanting (or 40 d after sowing)—would make a difference in
yields. According to Festinger’s (1957) cognitive dissonance theory, an information
in direct conflict with current beliefs can create a state of psychological dissonance
and can motivate information recipients to seek resolution through re-evaluation. The
potentia1 rewards when the question is resolved can serve as important incentives.
Because the use of conflicting information on changing perceptions works well when
information can be evaluated objectively (de Bono 1970), communicating new information
along with a simple farmer’s experiment for evaluation seems to be effective.
Training through the farmer field school (FFS) approach can also change perceptions
and pest management practices. In Indonesia, FFS-trained farmers reduced insecticide
spraying from an average of 2.8 sprays farmer -1 to 0.9 and the proportion of
farmers not spraying increased from 26% to 50% (Matteson et a1 1994, van de Fliert
et al 1995). In 15 wk of training, farmers often went through exercises in identifying
components of the ecosystem, discussing their functions (ecosystem analysis), and
doing a series of simple experiments to evaluate ideas. To evaluate whether leaf-feeding
insects would cause yield losses, farmers observed an experiment in which leaves
were removed by cutting and measured differences in yields.
An important role of research is to transform farmers’ needs into researchable
problems and communicate the results back to farmers (Pray and Echeverria 1990).
To facilitate this, we need to adopt a radical change in the research approach by emphasizing
understanding farmers’ pest management knowledge and practices and the
root causes of farmers’ problems. This has been discussed by Norton and Mumford
(1993), Bentley and Andrews (1996), and Heong (1996). We also need to emphasize
the participatory process, which begins with a problem definition from farmers’ perspectives
before setting research priorities and activities (Fig. 2). Even when research
is well targeted, the results may not get through to be adopted by farmers—a problem
of delivery (Norton and Mumford 1993). Disseminating, receiving, understanding,
internalizing, and using research information in decision making would probably require
improvements in communication. Because farmers live in an environment that
A comparative analysis of pest management practices of rice farmers in Asia 239

Fig. 2. A proposed radical change in the
research sequence. (Adapted from
Bentley and Andrews 1996.)
contains numerous competing sources of information, we need research to address
how communication media shape perceptions and attitudes of rice farmers. In pest
management, where risk aversion among farmers seems to be more pervasive, understanding
the impact of information provided by various sources on decision making
would help identify intervention opportunities.
References
Adjid DA. 1983. Vertical transfer of agrotechnology in Indonesia: the case of the BIMAS
Programme. In: Cagauan BG et al, editors. Agrotechnology transfer in Indonesia and the
Philippines. BSP Tech. Rep 9. IITAHR Res Series 027. Manoa (Hawaii, USA): University
of Hawaii. p 3-10.
Barker R, Bennager E, Hayami Y. 1978. New rice technology and policy alternatives for food
self-sufficiency. In: Economic consequences of the new rice technology. Los Baños (Philippines):
International Rice Research Institute. p 337-356.
Bentley J, Andrews K. 1996. Through the road blocks: IPM and Central American smallholders.
Gatekeeper Series No. 56. London (UK): International Institute for Environment and Development.
Bentley JW. 1989. What farmers don’t know can’t help them: the strengths and weaknesses of
indigenous technical knowledge in Honduras. Agric. Human Values 6:25-31.
Cohen JE, Schoenly K, Heong KL, Justo H, Arida G, Bamon AT, Litsinger JA. 1994. A food
web approach to evaluating the effect of insecticide spraying on insect pest population
dynamics in a Philippine irrigated rice ecosystem. J Appl. Ecol. 31:747-763.
Conway GR, Barbier ED. 1990. After the Green Revolution: sustainable agriculture for development.
London: Earthscan Publications.
Conway GR, Pretty JN. 1991. Unwelcome harvest: agriculture and pollution. London: Earthscan
Publications.
Cramer HH. 1967. Plant protection and world crop production. Pflanzen Schutz-Nachrichten
Bayer 20:1-254.
240 Heong and Escalada

David CC, Otsuka K, editors. 1994. Modern rice technology and income distribution in Asia.
Boulder (Colo., USA): Lynne Rienner Publishers, Inc.
De Bono E. 1970. Lateral thinking. Hermondsworth (Middesex, England): Penguin Books.
Fabella LT, Fabellar N, Heong KL. 1994. Simulating rice leaffolder feeding effects on yield
using MACROS. Int. Rice Res. Newsl. 19(2):7-8.
Festinger L. 1957. A theory of cognitive dissonance. Stanford (Calif., USA): Stanford University
Press.
Heinrichs EA, Mochida O. 1984. From secondary to major pest status: the case of insecticideinduced
rice brown planthopper, Nilaparvata lugens, resurgence. Protect. Ecol. 7:201-
218.
Heong KL. 1993. Rice leaffolders: are they serious pests? In: Research on rice leaffolder management
in China. Proceedings of the International Workshop on Economic Threshold for
Rice Leaffolders in China, 4-6 March 1992, Beijing, China. Hangzhou (P.R. China): National
Rice Research Institute. p 8-11.
Heong KL, Escalada MM. 1997. Perception change in rice pest management: a case study of
farmers’ evaluation of conflict information. J. Appl. Commun. 81(2):3-17.
Heong KL, Escalada MM, Lazaro AA. 1995a. Misuse of pesticides among rice farmers in
Leyte, Philippines. In: Pingali, PL, Roger PA, editors. Impact of pesticides on farmers’
health and the rice environment. Norwell (Mass., USA): Kluwer Academic Publishers.
p 97-108.
Heong KL, Ho NK, Jegatheesan S. 1985. The perception and management of pests among rice
farmers in the Muda Irrigation Scheme, Malaysia. MARDI Report No. 105. Kuala Lumpur
(Malaysia): Malaysian Agricultural Research and Development Institute. 11 p.
Heong KL, Teng PS, Moody K. 1995b. Managing rice pests with less chemicals. Geojournal
35:337-349.
Heong KL, Escalada MM, Vo Mai. 1994. An analysis of insecticide use in rice: case studies in
the Philippines and Vietnam. Int. J. Pest Manage. 40:173-178.
Heong KL. 1996. Pest management in tropical rice ecosystems: new paradigms for research.
In: Hokyo N, Norton G, editors. Proc. Int. Workshop on Pest Management Strategies in
Asian Monsoon Agroecosystems. Japan: Kyushu National Agricultural Experiment Station,
Ministry of Agriculture, Forestry and Fisheries. p 139-154.
IRRI (International Rice Research Institute). 1990. Crop loss assessment in rice. Los Baños
(Philippines): IRRI.
IRRI (International Rice Research Institute). 1994. Integrated pest management: the IRRI perspective.
IRRI Information Series No. 3. Los Baños (Philippines): IRRI.
IRRI (International Rice Research Institute). 1995. World rice statistics 1993-94. Los Baños
(Philippines): IRRI.
Kenmore PE, Cariño FO, Perez CA, Dyck VA, Gutierrez AP. 1984. Population regulation of
the rice brown planthopper (Nilaparvata lugens Stal.) within rice fields in the Philippines.
J. Plant Protect. Trop. 1:19-38.
Kenmore PE, Litsinger JA, Bandong JP, Santiago AC, Salac MM. 1987. Philippines rice farmers
and insecticides: thirty years of growing dependency and new options for change. In:
Tait J, Napompeth B, editors. Management of pests and pesticides—farmers’ perceptions
and practices. Boulder (Colo., USA): Westview Press. p 98-108.
Lim GS, Heong KL. 1984. The role of insecticides in rice integrated pest management. In:
Judicious and efficient use of insecticides on rice. Los Baños (Philippines): International
Rice Research Institute.
A comparative analysis of pest management practices of rice farmers in Asia 241

Litsinger JA, Price EC, Herrera RT. 1980. Small farmer pest control practices for rainfed rice,
corn and grain legumes in three Philippine provinces. Philippine Entomol. 4:65-80.
McKenzie W. 1996. World rice pesticide market. London: Wood McKenzie Consultants, Ltd.
Matteson PC, Gallagher KD, Kenmore PE. 1994. Extension of integrated pest management for
planthoppers in Asian irrigated rice: empowering the user. In: Planthoppers: their ecology
and management. London: Chapman & Hall. p 656-685.
Mumford JD, Norton GA. 1984. Economics of decisionmaking in pest management. Annu.
Rev. Entomol. 29:157-174.
Norton GA, Mumford JD, editors. 1993. Decision tools for pest management. Oxford (UK):
CAB International.
Panda N, Khush GS. 1995. Host plant resistance to insects. Wallingford (UK): CAB International.
Pathak MD, Khan ZR. 1994. Insect pests of rice. Los Baiios (Philippines): International Rice
Research Institute.
Pray C, Echeverria R. 1990. Private sector agricultural research and technology transfer links
in developing countries. In: Kaimowitz D, editor. Making the link: agricultural research
and technology transfer in developing countries. Boulder (Colo., USA): Westview Press.
p 197-216.
Rola AC, Pingali PL. 1993. Pesticides, rice productivity and farmers’ health: an economic
assessment. Los Baiios (Philippines): International Rice Research Institute.
Rombach MC, Gallagher KD. 1994. The brown planthopper: promises, problems and prospects.
In: Heinrichs EA, editor. Biology and management of rice insects. New Delhi (India):
Wiley Eastern. p 693-709.
Schoenly KG, Cohen JE, Heong KL, Arida GS, Barrion AT, Litsinger JA. 1994. Quantifying
the impact of insecticides on food web structure of rice arthropod populations in a Philippine
farmer’s irrigated field: a case study: In: Polis GA, Wisemiller K, editors. Food webs:
integration of patterns and dynamics. London: Chapman and Hall. p 343-351.
van de Fliert E, Pontius J, Roling N. 1995. Searching for strategies to replicate a successful
extension approach: training of IPM trainers in Indonesia. Eur. J. Agric. Educ. Exten.
1141-63.
Viajante V, Heinrichs EA. 1986. Rice growth and yield as affected by the whorl maggot, Hydrellia
philippina Ferino (Diptera: Ephydridae). Crop Protect. 5:176-181.
Way MJ, Heong KL. 1994. The role of biodiversity in the dynamics and management of insect
pests of tropical irrigated rice—a review. Bull. Entomol. Res. 84:567-587.
Notes
Authors’ addresses: M.M. Escalada, Department of Development Communication, Visayas State
College of Agriculture, Baybay, Leyte, Philippines; K.L. Heong, Entomology and Plant
Pathology Division, International Rice Research Institute, Los Bafios, Laguna, Philippines.
Acknowledgments: The authors would like to thank the Swiss Agency for Development and
Cooperation, through the Rice IPM Network based at the International Rice Research
Institute, for funding most of the farmer surveys reported, and all the authors who provided
us with the data, which served as the basis for this synthesis paper.
Citation: Heong KL, Escalada MM, editors. 1997. Pest management of rice farmers in Asia.
Manila (Philippines): International Rice Research Institute.
242 Heong and Escalada

Appendix I. A selection of local pest management terms used by Asian farmers
and their meanings.
Language Local terms Meanings
Pesticides
Khmer thnam poul Poisonous chemicals
Chinese nongyao Farm medicine
Indonesian obat Medicine
Lao san che mi Chemicals
Malaysian ubat Medicine
Myanmar po that say Medicine to kill insects
Philippines
Cebuano medisina Medicine
llocano agas Medicine
Tagalog kemikal Chemicals
Sinhala thel Oil or medicine
Tamil poochi marundhu Medicine to kill pests
Thai ya Chemicals for protection from
and eradication of crop pests
(including insects and diseases)
Vietnamese thuoc tru sau Medicine to kill worms
Insecticides
Khmer
Chinese
Hindi
Indonesian
Lao
Malaysian
Myanmar
Philippines
Cebuano
llocano
Tagalog
Sinhala
Tamil
Thai
Vietnamese
thnam poul
shachongyao
keet nashi dawa
obat
yaka maeng mai
ubat
po that say
medisina
agas ti igges
pamatay kulisap
mehi thel
poochi marundhu
ya kha malaeng
thuoc tru sau
Poisonous chemicals
Medicine to kill worms
Medicine to kill insects
Medicine
Medicine to kill insects
Medicine
Medicine to kill insects
Medicine
Medicine to kill worms
Insect killer
Medicine to kill insects
Medicine to kill insects
Medicine to kill insects
Medicine to kill worms
Fungicides
Khmer thnam samlab phaset Chemical for diseases
Chinese zhibingyao Medicine for treating disease
Hindi kawak nashi dawa Medicine to kill fungi
A comparative analysis of pest management practices of rice farmers in Asia 243

Appendix I continued.
Language Local terms Meanings
Indonesian
Lao
Malaysian
Myanmar
Philippines
Cebuano
llocano
Tagalog
Sinhala
Tamil
Thai
Vietnamese
obat
yaka phahad
ubat
mho that say
medisina
agas ti sakit
pampuksa ng sakit
thel
poonchala marundu
ya roke
thuoc tru benh
Medicine
Medicine against diseases
Medicine
Medicine to kill fungi
Medicine
Medicine for diseases
Disease killer
Oil or medicine
Medicine against diseases
Medicine against diseases
Medicine against diseases
Herbicides
Khmer
Chinese
Hindi
Indonesian
Lao
Malaysian
Myanmar
Philippines
Cebuano
llocano
Tagalog
Sinhala
Tamil
Thai
Vietnamese
thnam smao
chucaoji
kharpatwar nashi
obat rumput
yakayab
ubat numpai
pacing that say
medisina
medisina
pamatay damo
walnasaka
kalai kolli
ya kha yah
thuoc tru co
Chemicals for grasses
Material to remove grasses
Medicine to kill weeds
Medicine for grasses
Medicine to kill grasses
Medicine for grasses
Medicine to kill weeds
Medicine
Medicine for grasses
Medicine for grasses
Grass killer
Weed killer
Medicine to kill grasses
Medicine to kill grasses
Crop protection/plant protection
Khmer
Hindi phasal surakshya Crop protection
Indonesian perlindugan tanaman Crop protection
Lao pongkane maeng mai Pest control
Malaysian
perlindungan tanaman Crop security
Myanmar A pin poe mhar kar Plant protection from all pests
kaw yei
244 Heong and Escalada

Appendix I continued.
Language Local terms Meanings
Philippines
Cebuano
Tagalog
Sinhala
Tamil
Thai
Vietnamese
pagpanalipod sa tanum
peladi sanrakshana
kubkum peed
bare muamang
Plant protection
Plant security
Crop protection
Crop protection
Pest control
Khmer
Chinese
Hindi
Indonesian
Lao
Malaysian
Philippines
Cebuano
llocano
Tagalog
Sinhala
Tamil
Thai
Vietnamese
zhichong
niyantran
pemberantasan hama
ponkane maeng mai
kawal
pagsumpo sa dangan
krumi nasaka
kha malang
tru
Treat for worms
Control
Suppress insects
Control
Stop pests
Kill insects
Kill insects
Kill insects
A comparative analysis of pest management practices of rice farmers in Asia 245