1M5NBEg

ISSUE: 10JUNE, 2015CHILDSEXUALABUSEMATERIALSTRAFFICKING OFCHILDRENFOR SEXUALPURPOSESSEXUALEXPLOITATIONOF CHILDRENIN TRAVEL ANDTOURISMCHILDPROSTITUTIONRESEARCHINGTHE SEXUAL EXPLOITATIONOF CHILDREN:CHALLENGES AND METHODOLOGIES OF DATA COLLECTION

Series Editor: Rebecca H. RittenhouseECPAT would like to thank the following people for their review of the articles included in thisJournal issue 10:• Dr. Simon Baker• Rebecca Meiksin• Dr. Afua Twum-Danso ImohLayout & Design: Manida NaebklangJune, 2015Copyright © 2015, ECPAT InternationalThis publication was produced with the financial assistance of Irish Aid, Oak Foundation, and Sida.The views expressed herein are solely those of the authors. The support received from the abovementioneddonors does not constitute endorsement of the opinions expressed.ECPAT International328/1 Phayathai Road RatchathewiBangkok 10400 THAILANDTel: +662 215 3388, +662 611 0972 Fax: +662 215 8272Email: info@ecpat.net Website: www.ecpat.net

TABLE OF CONTENTSPREFACE 2TACKLING THE DATA DEARTH: THE GLOBAL SCALE OFCOMMERCIAL SEXUAL EXPLOITATION OF CHILDREN 4Introduction 4Reasons for the current lack of reliable CSEC-related data 5Commonly used global estimates of the scale and scope of CSEC 6Research approaches and the use of mixed-method data collection 7Introducing better evidence-based methods into CSEC research 8Conclusion 10Bibliography 11CONCEPTS AND PRINCIPLES RELATED TO QUALITY RESEARCHON THE COMMERCIAL SEXUAL EXPLOITATION OF CHILDREN 13Introduction 13Defining ‘quality research’ on CSEC 14Basic principles and concepts 15Conclusion 19Bibliography 20ETHICAL ISSUES OF RESEARCHING SEXUAL EXPLOITATION OF CHILDRENWITH VICTIMS, SURVIVORS AND THOSE AT RISK 22Introduction 22Is the involvement of children the best option? 24Protection of child participants from harm 25Context of the victim/survivor experience 26Selection of child participants and researchers 27Informed consent 27Addressing power imbalances 29Psychosocial support 30Right to privacy and confidentiality within a child protection framework 31Sharing the research outcomes 32Conclusions 32Bibliography 33AUTHOR BIOGRAPHIES 36

ISSUE : 10PREFACEThe lead-up to the 20th anniversary of the First World Congress against Commercial SexualExploitation of Children in 2016 is a perfect opportunity for the international community tostop and take stock of the progress made and the challenges still pending in ending the sexualexploitation of children. In order to do this, it is necessary – as a first step – to analyse what isknown at the national, regional and international levels about the sexual exploitation of childrenand the reliability and utility of this information.While focus on the sexual exploitation of children has increased over the last several decades,there remains limited reliable, evidence-based research into the depth, breadth and scope of thephenomenon. Organisations working to combat the commercial sexual exploitation of children(CSEC) are constantly requested to provide data and information on victims affected by theprinciple manifestations of CSEC (trafficking of children for sexual purposes, child prostitution,child pornography/child sexual abuse materials and the sexual exploitation of children in traveland tourism). However, it has been generally recognised that widely circulated figures on theglobal scale and magnitude of these violations against children remain ambiguous or lacking, inpart due to conceptual and methodological challenges. When conducting research on the sexualexploitation of children many issues should be considered, including not only the basic conceptsand principles relating to quality research but the importance of including the voices of childrenand the ethical concerns inherent in this approach.The reliance on limited, inaccurate and out-of-date data can have an adverse effect on advocacy,policy planning and targeted interventions related to the prevention and protection of childrenagainst sexual exploitation. In a 2013 report to the Human Rights Council, the Special Rapporteuron the sale of children, child prostitution and child pornography, Najat Maalla M’jid, emphasisedthat “the collection and analysis of reliable data on the sale and sexual exploitation of childrenremains a major challenge. The lack of reliable data reduces the visibility of the issue and thedevelopment of adequate responses and prevention.” 1ECPAT Journal Series No. 10 seeks to highlight this problem by analysing the methodologies andchallenges of reliable data collection when researching the sexual exploitation of children andsuggests possible ways to address these issues.The first article, Tackling the Data Dearth: the global scale of commercial sexual exploitationof children, analyses the lack of reliable studies and data on CSEC and the reasons for this. Themajor sources of existing information on the global scale and scope of CSEC are identified andexamined, and methods of producing more reliable data are discussed. The article concludes thatmore scientific and evidence-based research is critical in order to more effectively combat theproblem.1 UN Special Procedures, “Report of the Special Rapporteur on the sale of children, child prostitutionand child pornography, Najat Maalla M’jid”, UN Doc. A/HRC/25/48, (2013), para. 95.2

ISSUE : 10TACKLING THE DATA DEARTH:The global scale of commercial sexualexploitation of childrenby Jordana Dawson Hayes and Mark CapaldiCHILD PROSTITUTIONCHILDSEXUALABUSEMATERIALSTRAFFICKING OFCHILDRENFOR SEXUALPURPOSESSEXUAL EXPLOITATIONOF CHILDREN IN TRAVEL ANDTOURISMINTRODUCTIONThere is little doubt that the commercial sexual exploitation of children (CSEC) is a large andwidespread problem. However, while great strides have been made to tackle the issue, the needto introduce and utilise methodologically sound ways to define and measure this form of childabuse is becoming increasingly apparent. It has been recognised for over a decade that widelycirculatingfigures on the global scale of CSEC are frequently little more than “guesstimates” that,through frequent repetition, have attained the status of fact (Huijsmans and Baker, 2012; ECPATInternational, 2008).4

June, 2015Governments, donors and the public are now increasingly demanding evidence-based policiesand thorough monitoring and evaluation techniques to demonstrate efficacy in the fight againstCSEC. It is clear that a “shift [is] happening, a shift away from glossy brochures and smiling childrenand happy anecdotes, a shift toward data” (Kestenbaum, 2013). Not only do child rights agenciesneed to be able to efficiently allocate limited resources, maintain funding and evaluate theirown work for the benefit of their target populations, investors and donors need to be able todemonstrate the impact of their charitable “investments” (for marketing purposes, for example).Inaccurate and inflated numbers may discourage efforts to address the problem of exploitation(Feingold, 2010) or cause duty-bearers to promote less effective policies and projects. In theworst situations, research with major methodological flaws can harm those it intends to help,by stigmatising or putting victims at risk, or by leading decision-makers to inaccurate conclusions(SIREN, 2011).This article seeks to examine some of the issues relating to data collection in the area of thecommercial sexual exploitation of children. The major sources of existing information on the globalscale and scope of CSEC are identified and examined, and methods to produce more reliable dataare discussed. The article concludes that more scientific and evidence-based research on CSEC iscritical in order to more effectively combat this phenomenon.REASONS FORTHE CURRENTLACK OF RELIABLECSEC-RELATEDDATAA quick Internet search willreveal to the interestedresearcher the discrepanciesthat exist among CSEC statistics,as well as the lack of adequatecitations and references toreliable sources. Several reasonsare often given for the lack ofreliable data on CSEC, the mostprominent of which is that childvictims of sexual exploitationconstitute a “hidden population”that is inherently challengingto research due to the difficultyin finding a representativesample. While this is true, thisproblem is not unique to CSEC.Research in a multitude of otherfields such as adult prostitution(Paz-Bailey, et al., 2011) andillegal drug abuse (Dombrowski,et al., 2012; Salganik, et al.,2011), demonstrates that thereare continuously improvingmethodological and scientificapproaches to work around thisbarrier.In addition to the hidden natureof CSEC, other barriers shouldbe addressed before data canbe analysed to achieve the mostreliable results. Global data onsexual exploitation is limited,but the problem is exacerbatedfor CSEC because figures forwomen and children are oftencombined (ECPAT International,2008; Kelly, 2002). Datacollected by researchers needsto be effectively disaggregatedto achieve the most out of thedata. Different interpretations ofinternationally recognised legaldefinitions of related aspectsof CSEC have compounded thedifficulties in disaggregating dataor resulted in little informationon particularly vulnerablegroups such as the sexualexploitation of transgenderchildren and adolescentboys (MenCare, 2012; ECPATInternational, 2012).Several reasons areoften given for thelack of reliable dataon CSEC, the mostprominent of whichis that child victimsof sexual exploitationconstitute a “hiddenpopulation” thatis inherentlychallenging toresearch due to thedifficulty in findinga representativesample5

ISSUE : 10The commercial sexualexploitation of children iscomprised of several separatebut related manifestationsof child sexual abuse – childprostitution, child pornography/child sexual abuse materials,sexual exploitation of childrenin travel and tourism andtrafficking of children forsexual purposes – and studiesoften focus on one specificmanifestation rather than onCSEC in its entirety. Methods,definitions, indicatorsand research quality varysignificantly among thesestudies which then make themdifficult to compare or compile(Weitzer, 2014).Other research problems includethe fact that research plansand methodologies are oftennot made public, which limitspeer-reviewability (Huijsmansand Baker, 2012; ECPATInternational, 2008). There isfrequent sampling bias, minimaluse of controls and, mostResearchmethods notmade publicSeveral separatebut relatedmanifestationsof CSECsignificantly, little admissionof these flaws in the publishedresearch. At the secondaryresearch level, there areoften examples of poor citingand source checking amongorganisations and minimal useof peer-reviewed and verifiablereferences (Weitzer, 2014). Thefollowing section of this paperexamines current global CSECestimates and exemplifies someof these problems.COMMONLY USEDGLOBAL ESTIMATESOF THE SCALE ANDSCOPE OF CSECThe global scale of CSECestimates that do exist appearto stem from statistics providedby the International LabourOrganization (ILO). The two mainfigures frequently used are thatHiddenpopulationLack ofdisaggregateddata1.8 million children are exploitedworldwide through prostitutionand pornography, and that 1.2million children globally areinvolved in trafficking. The ILO(n.d.) states that these figurescome from its 2000 child labourdata which was gathered from“various secondary sources.”A UNICEF report that referredto the 1.2 million ILO figureis also frequently referencedas a data source (UNICEF,2006) – although it is oftenerroneously cited as statingthat 2 million children arevictims of commercial sexualexploitation, including bythe ILO (International LabourOrganization - InternationalProgramme on the Eliminationof Child Labour, 2008). 1In 2005, ILO estimated that1.39 million people arevictims of commercial sexualexploitation at any given timewith 40-50% of these individualsbeing minors, putting thenumber of minors around 0.79million (International LabourOrganization, 2005). Moredetailed information on howthese figures were arrived at(as well as methodologicallimitations) were provided inthe 2005 report, unlike previousILO reports, suggesting that the1.39 million figure is the closestto a reliable global CSEC-relatedfigure existing today.These challenges are not uniqueto the international contextand similar dubious figurescan be seen at the nationallevel where studies also lackempirical data to support thenumbers given (Weitzer, 2014).However, organisations workingto support and advocate for1.UNICEF quotes the “2 million figure” in its own press release of the report (http://www.unicef.org/sowc06/press/release.php).6

June, 2015victimised youth often feelcompelled to use statistics –even if the numbers are lessthan reliable – because largernumbers can be importantawareness-raising and advocacytools (ECPAT International,2014). At the same time, datacollection and analysis can seeminsignificant when compared tothe primary goal of protectingchildren from such a horrificcrime, and organisations maybe of the opinion that devotingefforts to more reliable datacollection and analysis is a pooruse of resources or a distraction(SIREN, 2011).RESEARCHAPPROACHES ANDTHE USE OF MIXED-METHOD DATACOLLECTIONThe current dearth of relevantdata and the problems withexisting reliable data allowconclusions to be drawn as towhat research into the scale andscope of CSEC should ideallyinclude. Preferably, researchapproaches need to be ableto measure prevalence aswell as to determine ways toidentify and mitigate against thevulnerabilities and hazards thatchildren at risk or involved incommercial sexual exploitationface (SIREN, 2011). Greaterattention is needed to ensuremethodological and ethical rigorin CSEC research. As no researchmethodology is completelysuperior or more definitively ofuse, the final selection of themost appropriate methodologyshould be related to the purposeof the research and the level ofdescription and interpretationrequired (Sandelowski, 2000).Qualitative researchmethodology has long beenthe most common approachof analysis when detaileddescriptions of a phenomenonare required as it allows therespondents to describe theirexperiences in their own words(Wicks and Whiteford, 2006).Methods such as life historynarratives or case studiesare particularly useful whenresearching with adolescentsas older children have thecapacity to think conceptuallyand can be interested incontributing to meaningfulconversations where they canpresent their experiences andviews (Haglund, 2004). Donewell, such qualitative researchapproaches provide uniquestandards of reliability andvalidity on the meaning andexperiences as lived by thesechildren which transcendconcerns of representation.Critics of qualitative researchpoint to the generally small andunscientific sampling techniquesThe currentdearth ofrelevant data andthe problemswith existingreliable dataallow conclusionsto be drawn asto what researchinto the scale andscope of CSECshould ideallyincludeoften used with qualitativeresearch. Frequent examplesin CSEC-related researchinclude unrepresentativesamples; problems with accessto respondents; selectionbiases by “gate-keepers” torespondents such as NGOs,social workers and police; and –where participants are referredby other participants – biasedselection of participants basedon friendships or other similarfactors (Brunovskis and Surtees,2010). This also leads to theinability to scale up estimatesto the national or global level asthe research may not be trulyrepresentative of the widertarget population (Brunovskisand Surtees, 2010; Morgan,2008).Quantitative research typicallyutilises surveys or otherpre-structured methods toobtain a more representativedataset on pre-selectedvariables. Statistics are usedto interpret the findings. Theresearcher generally pre-selectsthe variables to be studiedwithin pre-set confines andconcepts (Sandelowski, 2000).Quantitative research, whilemeeting high standards ofrepresentation and scientificallysoundanalysis, can limitwhat can be learned aboutthe meanings intervieweesgive to certain events andtherefore leave little roomfor the unexpected (Becker,1996). With these challengesin mind, quantitative research(often utilising surveys orquestionnaires) must first betechnically strong. Researchneeds to be methodologicallysound and must be replicableacross countries and situations.Bias should be eliminated tothe extent possible and anyremaining bias and research7

ISSUE : 10limitations must be describedand discussed in the resultingpublication. In determiningsample size, quantitativeapproaches must be designed insuch a way that the research: (i)covers a representative part of aclearly defined target population(i.e., child victims of commercialsexual exploitation); and (ii)identifies members of the targetpopulation and distinguishesthem from non-members(Tyldum, 2009).Mixed methods utilising bothqualitative and quantitativeresearch are particularly helpfulin exploring conceptsthat are complex and multifaceted(Capaldi, 2014; Woodand Welch, 2010) and theyenable the triangulation oftheoretical constructs anddata from generally largersample sizes. Integrating anddrawing conclusions from bothquantitative and qualitativeresearch methods respondsto criticisms of reduced vigourand validity, enabling betterestimations of the scale andscope of hidden crimes suchas the sexual exploitation ofchildren (Wood and Welch,2010; Habashi and Worley,2009).INTRODUCINGBETTER EVIDENCE-BASED METHODSINTO CSEC RESEARCHWhen estimating the magnitudeof any illicit activity such asCSEC, the need for qualitydata collection and verifiablesources to arrive at the figuresis paramount. The continuingdisparity among the numbers isleading to many organisationsand academics casting doubt onthe efficacy of producing globallevel estimates (Weitzer, 2014;Zhang, 2012).Work through alocal communityRespondentdrivensamplingStratified random sampling8

June, 2015There are several promisingapproaches to estimating thesize of a hidden population. TheStrategic Information ResponseNetwork (SIREN), based inSoutheast Asia, recentlyorganised a competition forgood practice data gathering inthe field of human trafficking.The competition formatappeared to identify severalgood ideas that, through furtherinvestigation and collaboration,could be applied to researchon the sexual exploitationof children. For example,participants at the Universityof Miami in the United Statesdescribed a communicationbasedmethod that wouldenable researchers to workthrough a local community,rather than law enforcementsystems, thereby avoiding thecommon problem of having toextrapolate up a multiplicationfactor for reported cases ofCSEC. A second advantageis that the method does notrely on relationships betweenparticipants (a common trendin hidden population researchdesigns), which may be anywaysmaller in child populations.The method assumes thatthe commercial sex traderequires knowledge of supplyand demand patterns andthat local informants providelocation information to potentialcustomers and in return receivefinancial remuneration. Thus,this methodology seeks tolocate sex trade venues andinformation through localinformants (SIREN, 2008).Triangulation of a broad rangeof stakeholders is also currentlybeing piloted by UNICEF, inpartnership with nationalgovernments and the Centrefor Disease Control, but in thisapproach quantitative surveys inmulti-country studies are usedto examine violence againstchildren (UNICEF, 2011).A second approach – stratifiedrandom sampling – also avoidsthe tendency to work throughlaw enforcement agencies andacquaintance circles as thenumber of assisted victims willnot be similar to the estimatednumber of victims (Weitzer,2014). Stratified samplinginvolves dividing members ofa population into homogenousgroups before sampling. Whenused to survey a town forexample, it can ensure thatdiversity within the populationis reflected and can control forpopulation density variancewithin the area (SIREN, 2008).Respondent-driven sampling(RDS) is a technique thatdemonstrates improvementin the more traditional linktracing(snowball) design whererespondents recruit futurerespondents from their circle ofacquaintances. RDS improvesupon snowball sampling byallowing researchers to makeasymptotically unbiasedestimates from snowballsamples under certainconditions (Fisher and GiovannaMerli, 2014). However, oneof the primary conditions ofRDS is that researchers shouldnot attempt to estimatedirectly from the sample to thepopulation (Salganik, 2006);hence, RDS’s most obvious useis to estimate the prevalenceof a specific trait among thesample. However, there is atleast one novel example ofRDS being used to determinesample size (Handcock andMar, 2012). These researchershave presented an approachto estimating the size of atarget population basedRespondent-drivensampling (RDS) isa technique thatdemonstratesimprovement in themore traditionallink-tracing(snowball) designwhere respondentsrecruit futurerespondents fromtheir circle ofacquaintances.on data collected throughRDS that uses a “successivesampling approximation toRDS to leverage information”in harder to reach samplesby concentrating on the datacollected from the largersamples accessed first. This isan important development inhidden-population research andshould not be overlooked byCSEC researchers.Nevertheless, making macrolevelclaims on the scale ofCSEC or human trafficking isstill frought with challenges.National level surveysoffer possibilities for morestandardised data collectionand counting but a trulyrepresentative national surveyof victims is still extremelychallenging due to the hiddenpopulation whose boundariesremain largely unknown.Therefore, surveys with a largerpopulation that is amenableto probability-based samplingthrough association mayproduce promising results(Weitzer, 2014). For example,the International Organization9

ISSUE : 10for Migration (IOM) conductedmulti-country national researchon human trafficking in fivehigh-risk European countries,sampling 5513 randomlyselected households. Thosesurveyed were not askedabout their own experiencesof being trafficked but whethera close family member hadbeen trafficked, with 2% ofthe sample reporting familymembers who fit the definitionof a victim, according to thesurvey (Omae Mahmoudand Trebesch, 2010). Suchmethods are rarely used andthere remains potential biasdue to family sensitivities andstigmatisation related to illicitactivities such as CSEC (Weitzer,2014).CONCLUSIONIt is clear that there remains apervasive lack of quality andreliable data around the scaleand scope of the commercialsexual exploitation of childrenglobally and that figuresfrequently used are littlemore than “guesstimates.”This is a particular concernbecause quality data allowsorganisations to make thebest decisions on where todedicate limited resources andto objectively monitor andevaluate implemented projectsin terms of efficacy. It alsoallows donors to demonstratethe impact of their investments.Even with limited funds andvictims in urgent need, it isnecessary to spend resourceson rigorous research to ensureresources are correctly targetedand to increase effectivenessand efficiency of policies andprogrammes.All methods of data gatheringwill involve biases andlimitations and any large scaleproject to determine the scaleand scope of CSEC shouldinvolve a thorough analysis byresearchers and statisticians asto the method(s) to be utilised;collaboration between childrights agencies and academicsor research institutions areparticularly beneficial. Thesechallenges are not unique toCSEC or hidden populationresearch. Realising and reportingon biases and limitations isequally as important as utilisinga reliable and replicable surveydesign. Ability to disaggregatedata is especially important forresearch into specialised topics,such as CSEC, which are oftenviewed as subsets of a widerproblem.Mixed methodologicalresearch that integrates bothqualitative and quantitativetools provides the bestopportunities to produceempirical and scientific datathat is most representationalof the larger target population.Truly randomised samplingand large sample sizes canreduce biases and inaccuracies.Nevertheless, dangers existwhen extrapolating figuresup to national, regional andglobal levels and the significantresources required to undertakemacro-level research perhapsexplains why many globalestimates continue to beproblematic and inaccurate.The commercial sexualexploitation of children isa difficult area in which togather reliable data. However,it is crucial that organisationsworking to combat child sexualexploitation overcome thebarriers associated with thisresearch and commit to gatherevidence-based data. The datais essential to make the best useof limited funds, to set baselinesand targets, to produce themost efficient and targetedprogrammes and to effectivelymonitor and evaluate efforts toeradicate the commercial sexualexploitation of children.10

June, 2015BIBLIOGRAPHYAsquith, S. and E. Turner (2008). Recovery and reintegration of children from the effects of sexualexploitation and related trafficking. Geneva: Oak Foundation.Becker, H. (1996). “The epistemology of qualitative research”. In R. Jessor, A. Colby and R. Shweder, eds.Ethnography and human development: context and meaning in social inquiry. Chicago: University ofChicago Press.Brunovskis, A. and R. Surtees (2010). “Untold Stories: biases and selection effects in research with victimsof trafficking for sexual exploitation”. International Migration. Vol. 48, no. 4 (2010): 1-37.Capaldi, M. (2014). The child’s journey in search of rights: determining and addressing points of vulnerabilityin independent child migration in Thailand. Bangkok: Terre des Hommes Netherlands.Dombrowski, K., B. Khan, T. Wendel, K. McLean, E. Misshula, and R. Curtis (2012). “Estimating the sizeof the methamphetamine-using population in New York City using network sampling techniques”.Advances in Applied Sociology. Vol. 2, no. 4 (2012): 245-252.ECPAT International (2008). Exploitation of children in prostitution: Thematic Paper for the World CongressIII. Bangkok: ECPAT International.ECPAT International (2012). “Hidden violence: preventing and responding to sexual exploitation and sexualabuse of adolescent boys”. Compendium of articles: rights in practice in combating sexual exploitationof children, Journal Series No. 3. Bangkok: ECPAT International.ECPAT International (2014). Ending child prostitution, child pornography and trafficking for sexual purposes:unfinished business, External Scan. Bangkok: ECPAT International.Feingold, D. (2010). “Trafficking in numbers: the social construction of human trafficking data”. In P. A.eds. Sex, drugs and body counts: the politics of numbers in global crime and conflict. p.14. CornellUniversity Press.Fisher, J. and M. Giovanna Merli (2014). “Stickiness of respondent-driven sampling recruitment chains”.Network Science. Vol. 2, no. 2 (2014): 298-301.Habashi, J. and J. Worley (2009). “Child geopolitical agency: a mixed methods case study”. Journal ofMixed Methods Research. Vol. 3, no. 1 (2009): 42-64.Haglund, K. (2004). “Conducting life history research with adolescents”. Qualitative Health Research. Vol.14, no. 9.Handcock, M.G., K. Gile and C. Mar (2012). “Estimating hidden population size using respondent-drivensampling data”. Stat.ME. arXiv: 1209.6241v1.Huijsmans, R. and S. Baker (2012). “Child trafficking: ‘worst form’ of child labour or worst approach toyoung migrants?”. Development and Change. Vol. 43, no. 4 (2012): 919-946.ILO (n.d.). “Commercial Sexual Exploitation of Children and Adolescents: The I.L.O.’s Response”. Accessedon 30 November 2014 at: http://www.ilo.org/ipec/areas/CSEC/lang--en/index.htmInternational Labour Organization (2005). “A global alliance against forced labour: global report underthe follow-up to the ILO declaration on fundamental principles and rights at work”. Report I(B), ILOConference 93rd session 2005. pp.12,15. Geneva: ILO.International Labour Organization - International Programme on the Elimination of Child Labour (2008).Commercial sexual exploitation of children and adolescents: the ILO’s response. Geneva: ILO-IPEC.Kelly, L. (2002). “Journeys of jeopardy: a review of research on trafficking in women and children inEurope”. IOM Migration Research Series No. 11.11

ISSUE : 10Kestenbaum, D. (2013). “I was just trying to help”. Podcast. Chicago, Illinois, USA.MenCare (2012). Hidden violence: preventing and responding to sexual exploitation and sexual abuse ofadolescent boys. A Briefing Paper by MenCare.Morgan, D. (2008). The SAGE encyclopedia of qualitative research methods. pp. 816-817. SAGEPublications, Inc.Omae Mahmoud, T. and C. Trebesch (2010). The Economics of Human Trafficking and Labour Migration:Micro-Evidence from Eastern Europe. Accessed at: http://www.iza.org/conference_files/LeIlli2010/trebesch_c4269.pdfPaz-Bailey, G., J.O. Jacobson, M.E. Guardado, F.M. Hernandez, A.I. Nieto, M. Estrada and J. Creswell (2011).“How many men who have sex with men and female sex workers live in El Salvador? Using respondentdrivensampling and capture-recapture to estimate population sizes”. Sexually Transmitted Infections.Vol. 87, no. 4 (2011): 279-282.Salganik, M. (2006). “Variance estimation, design effects, and sample size calculations for respondentdrivensampling”. Journal of Urban Health. Vol. 83, no. 1 (2006): 98-112.Salganik, M., M.B. Mello, A.H. Abdo, N. Bertoni, D. Fazito and F.I. Bastos (2011). “The game of contacts:estimating the social visibility of groups”. Social Networks. Vol. 33, no. 1 (2011): 70-78.Sandelowski, M. (2000). “Whatever happened to qualitative description?”. Research in Nursing andHealth. Vol. 23 (2000): 334-340.SIREN (2008). Statistical methods for estimating numbers of trafficking victims. Bangkok, Thailand: UnitedNations Inter-Agency Project on Human Trafficking: Phase III.SIREN (2011). The state of counter-trafficking research: researcher, programmer and donor perspectives.Bangkok, Thailand: United Nations Inter-Agency Project on Human Trafficking: Phase III.Tyldum, G. (2009). “Coping with Biases in Trafficking Data”. In Human trafficking: new directions forresearch. International Organization for Migration.UNICEF (2006). The state of the world’s children: excluded and invisible. New York, USA.UNICEF (2011). Violence against children in Tanzania: findings from a national survey 2009. Tanzania:UNICEF, U.S. Centers for Disease Control and Prevention, Muhimbili University of Health and AlliedSciences.Weitzer, R. (2014). “New directions in research on human trafficking”. The ANNALS of the AmericanAcademy of Political and Social Science. Vol. 653, no. 6.Wicks, A. and G. Whiteford (2006). “Conceptual and practical issues in qualitative research: reflectionson a life-history study”. Scandinavian Journal of Occupational Therapy. Vol. 13, no. 2 (2006): 94-100.Wood, M. and C. Welch (2010). “Are ‘qualitative’ and ‘quantitative’ useful terms for describing research?”.Methodological Innovations Online. Vol. 5, no. 1 (2010): 56-71.Zhang, S. (2012). “Measuring labour trafficking: a research note”. Crime, Law and Social Change. Vol. 58,no. 4 (2012): 469-482.12

ISSUE : 10Organisations that are working to combat the commercial sexual exploitation of children (CSEC)are constantly requested to provide evidence-based data and information on the child victimsaffected by the various manifestations of sexual exploitation. However, accurate estimates,descriptions or evidence of the extent and magnitude of these violations against children remainambiguous or lacking, in part due to conceptual and methodological challenges (UNICEF, 2008).This results in limited, inaccurate and out-of-date knowledge on the phenomenon which has anadverse effect on advocacy, policy planning and targeted interventions (Weitzer, 2014; Huijsmansand Baker, 2012). Furthermore, poor quality research can even harm those it is intending tohelp through stigmatisation or putting victims at further risk (SIREN, 2011). When advocacy isbased on vague, non-scientific and misleading data it is not reliable and, as a consequence, itfails in achieving the goal of pushing duty-bearers to properly engage in addressing exploitation(Feingold, 2010).Quality research on a hidden phenomenon such as the sexual exploitation of children is complexand more variegated than many of the studies that currently exist. This article therefore distilssome of the core concepts and principles needed to undertake better quality CSEC research inthe hope that it can motivate key stakeholders to commit to more superior and evidence-basedstudies. It argues that much greater prioritisation needs to be given to methodological and ethicalrigour in research in order to maximise the pertinence and impact of the data collected and thus,to strengthen the implementation of effectual protection policies and interventions.DEFINING ‘QUALITYRESEARCH’ ON CSECResearch on commercial sexualexploitation of children is oftencriticised for being unscientific,non evidence-based and“unsubstantiated, misleading,exaggerated and sensational”(Ennew, 2008).In order to overcome suchchallenges and gaps in research,a major effort towards ‘qualityresearch’ is required. The qualityof research depends on thelevel of reliability (in scientificterms) and appropriateness ofresearch methods. This is oftenperceived as to do with the typeof information collected duringthe research, i.e. whetherit is numbers and statistics(quantitative data) or narrativesand images (qualitative data).Despite there not being ahierarchical relation betweenthese two types of information(as both are equally importantand mutually supportive),qualitative research is oftencriticised as less reliable and notscientific (Sandelowski, 2000).As regards CSEC, the bulk ofresearch on this issue – and onchildren in general – is mainlyqualitative research, whichmeans that policymakers andprogramme designers may nottake the information seriously.In order to change thisperception, it is necessaryto highlight the function andimportance of all methods ofresearch and to demonstratethat even qualitativeapproaches to data collectioncan be conducted in a scientificway and thus be sufficientlyreliable to inform policiesand programmes. In fact,quantitative and qualitativeThe quality ofresearch dependson the level ofreliability andappropriateness ofresearch methods.This is oftenperceived as todo with the typeof informationcollected duringthe research,i.e. whetherit is numbersand statistics ornarratives andimages.data are inter-related becauseit’s impossible to collectquantitative data without14

June, 2015first having a preliminaryknowledge of the subject basedon qualitative data of what thenumbers mean (Sandelowski,2000). Descriptions are thebasis of numbers, which meansthat qualitative data must befirst collected when there isweak knowledge about thephenomenon that is beingstudied. Although qualitativeapproaches tend to involvefewer participants comparedto quantitative research (whichis typically looking for trendsor estimating populationsizes), in qualitative studiesthe researchers generallyknow more details about eachparticipant and can thereforedevelop theories and newmeanings to describe or explainevolving realities of a complexphenomenon. This can becompared to quantitativeresearch where the participants’responses are recorded as‘numbers’, often in order toextrapolate or scale up data(Keele, n.d.).No one research methodology isnecessarily better than another.There is a need to developmore rigorous qualitative andquantitative research promotinga mixed methodology (Beije, etal., 2013) or ‘umbrella’ approachto research which recognisesthe different combinations ofapproaches to data collection.BASIC PRINCIPLESAND CONCEPTSA. CONCEPTUALANALYSIS ANDDEFINITIONSThe study of sexual exploitationof children cannot beadequately conducted ifdefinitions of the main conceptsrelated to the phenomenonand to the specific object of thestudy are not established soas to avoid misunderstandingsand vagueness of the dataitself. Defining theories,concepts and terminology isnot an unnecessary theoreticalexercise but a substantial needin research (Ennew, 2008).Theoretical and conceptualinaccuracies impede quality datacollection which then becomesover simplistic, insufficientlydisaggregated and rarelycomparable with other contextsor studies (Weitzer, 2014).A further weakness in thecurrent research on CSEC isthe almost total absence ofcontext analysis, in particular,as regards the local prostitutionmechanisms, economies, socialenvironment and culturalperceptions (such as ideasabout children and childhood,gender and sexuality and powerrelations). This has resulted inbroad generalisations aboutroot causes of CSEC, especiallyfamily breakdown, poverty, childmigration and social tolerance(ECPAT International, 2014).Conversely, it can also lead tothe compartmentalisation ofthe exploitation of childrenwithin either sex trafficking, sextourism or child pornography.In fact, these are phenomenathat beyond being interlinkedare also heavily inter-relatedwith and influenced by thecontexts in which they occur(Ennew, 2008). The lack ofcontext analysis in researchon CSEC prevents a holisticand reliable understanding ofthe complex nature of sexualexploitation of children and theinterplay between the variousmanifestations. Conceptualdifferences of distinct situationsof CSEC will inevitably result inthe need for different responses(ECPAT International, 2014).Greater awareness andknowledge conceptually ofthe nuances, distinctions andterminology around CSEC willhelp to strengthen the design,effectiveness and monitoringof counteraction (NGO Group,2005). Furthermore, clarityand analysis within the contextof conceptual frameworkson existing national andinternational legal frameworksalso needs to be provided, sothat the use and enforcementof legislation continues topromote and support globalharmonisation of disaggregateddata collection, cutting edgeresearch and good practices(ECPAT International, 2014).B. RIGHTS-BASEDAPPROACH TORESEARCHING WITHCHILDRENArticle 12.1 of the Conventionon the Rights of the Child makesit clear that children are notmerely ‘objects’ of concernand protection but ‘subjects’entitled to human rights amongwhich includes the right toexpress their views in all mattersconcerning them. Whilst the15

ISSUE : 10Convention does not specificallyrecognise it, children’s ‘right tobe properly researched’ maybe identified in it (KnowingChildren, 2009). This right,that children are indirectlyentitled to, is based on theinterpretation of Articles 3, 12,13 and 36 of the UN CRC whichsuggest direct implications inthe research about children. 1The meaningful participationof children in research canaccordingly improve the qualityand reliability of the findingsand if handled correctly canbe both therapeutic andempowering (Edmunds, 2003).As regards research on CSEC,the involvement of childrenand adolescents can occurat different levels and with avariable intensity. Children canparticipate in focus groups orinterviews as respondents,or it can be deeper and moreintense, such as in peer researchand youth-led studies in whichchildren and young peopleare involved as researchersthemselves (Akerstrom andBrunnberg, 2013). This meansthat the nature and level ofparticipation of children andyoung people in researchmay differ significantly, butthe principle on which theparticipatory approach is basedremains unchanged.It is evident that in carryingout research the participationof victims of CSEC raisessome extremely sensitiveethical issues. Research canbe exploitative in itself andparticipation is potentiallyharmful or dangerous forchildren and young peoplewho take part in the research(as discussed in the sectionbelow). For this reason, researchthat adopts a participatoryapproach has to use rightsbasedmethodology in order tobe meaningful and not harmfulfor the participants (Alderson,2004). As such, research withchildren must be based on andoriented by fundamental humanrights and the main principlesof dignity, equality, nondiscriminationand participation(Knowing Children, 2009).C. ETHICAL CONCERNSThe involvement of childrenand young people, in particularexperiential children and youth,in research on commercialsexual exploitation of childrenobviously raises major ethicalconcerns, such as whether thiskind of research is worthwhileor ‘fair’. One response might bethat despite all the ethical issuesraised by the participatoryapproach, it would also beunethical to do research onCSEC and deny children theright to be heard on somethingthat affects them (Laws andMann, 2004). However, seriousconsiderations need to be givento the ethical and protectionframework in participatoryresearch on CSEC.In order to prevent harmful orpoor quality data collection,risk-benefit assessments need tobe made at different levels: bythe researchers; by ethics andscientific review committees andadvisers; 2 by the people whoare asked to take part in theresearch (children and youngpeople) and, if relevant, theircarers (Alderson and Morrow,2004). In the planning of thestudy, it is important for theresearch team to explore themotivations for seeking toinvolve children in the project;the risks they face by embarkingon the project; and their abilitiesand responsibilities towardsthe children with whom theyare engaging (Harcourt andQuennerstedt, 2014).An ethical approach to researchwith children refers not onlyto traditional areas of concernlike voluntary participation andinformed consent, but to manyother aspects as shown in Table1 on the next page.1.See Article 3.3: “State parties shall……conform with the standards established by competent authorities”; Article 12: The childhas “….the right to express those views freely in all matters affecting the child”; Article 13.1: “The child shall have the rightto freedom of expression…….either orally, in writing or in print, in the form of art, or through any other media of the child’schoice”; and Article 36: Children must be protected “against all forms of exploitation” (Knowing Children, 2009).2.Ethical/Scientific/Advisory Committees generally consider three key criteria: scientific and methodological validity; the welfareof the participants; and the respect for the dignity and rights of participants (Allmark, 2003).16

June, 2015TABLE 1: ETHICAL PRINCIPLES WHEN RESEARCHING WITH CHILDRENPRINCIPLESVoluntary and meaningfulparticipation of childrenInformed consentProtection of children from anykind of harmRisks faced by the project teamduring the researchConfidentiality of data andprotection of respondents’interestsCompensatory expectations ofchildren and youth involved inthe researchProfessional skills, knowledgeand personal factors ofresearchersChild-friendly methods adoptedwith young respondentsISSUES TO CONSIDEREnsure no repercussions for respondent opting-out and thatthe child is not coerced into participating; child-centred andchild-friendly research methodologies should be utilised so thatthe child can fully and appropriately engage (Ansell, 2009).Consent is only viewed as valid if the person giving it issufficiently competent and informed to give their decision andit is made completely voluntarily as an on-going consideration(Alderson, 2004).No interview should take place if it is likely to put a child oryouth in a worse position, either in the short or long term;Access to referral mechanisms for child victims of sexualexploitation needing support should be available.As CSEC is a criminal activity, a risk management assessmentshould be undertaken to ensure that members of the researchteam do not face security risks or reprisals from criminalelements or corrupt officials in authority.Personal details such as real/full names, locations/addresses(or any other information which may help to identify the childrespondent) should be kept in a secure and private place andnot used in the research report. Researchers and translatorsshould sign a confidentiality agreement and respondentsshould be informed of the precautions being taken to protecttheir identity (Capaldi, 2014).There are pros and cons associated with the different types ofremuneration (reimbursement, compensation, appreciation orincentive) that may be given to a child respondent – explicitpolicies and guidelines justifying the approach used should bedeveloped (Alderson and Morrow, 2004).Persons undertaking research should be adequately trained.Researchers should be of relevant gender; have experience ofinterviewing children; be trained in victim support work; andhave at least a basic knowledge and understanding of childsexual exploitation (Bejerken, 2005).Create a comfortable environment for interviewees. Developa friendly and appropriate relationship of trust with therespondents (Kovačević and Mirović, 2007). Use a range ofparticipatory methodologies such as through ‘drawing andstories’, role play, photo essays, etc. (Knowing Children, 2009).17

ISSUE : 10PRINCIPLESImbalance of power betweenadult researcher and childrenResponsibilities/responses ofresearchers to children whoshow signs of distress or whoare currently being sexuallyexploitedOpportunity to return somethingback to children for theirparticipationISSUES TO CONSIDERBe aware that positionality and relations of power (due toage, gender, nationality, socio-economic aspects, etc.) canintimidate respondents or result in biases or prejudices in thedata collected (Ennew, 2008).Respect children’s ongoing right to refrain from answering anyquestion or withdraw from the research project if necessary.Facilitate access to available support should they feeluncomfortable and show signs of distress during and followingthe interview process. Explain limits to confidentiality andpossible implications such as appropriate referral should a childreport current abuse, including mandatory reporting in somecountries (Kovačević and Mirović, 2007).Through participation, CSEC victims/survivors can feelempowered, validated and actively heard by having theirinsights and concerns taken seriously; share research resultswith the respondents, including with children (Laws and Mann,2004).Although these principlesand guidelines are necessary,it must be noted that theycannot replace ‘contingentethics’ (Morrow and Richards,1996). Ethical guidelines areunlikely to provide specific,clear applications to all of thedilemmas that researchers face(Alderson, 1995). Researchersneed to be aware that ethicalconsiderations are ongoing andthat ethical dilemmas may ariseat any stage of the research.Decisions must be made inspecific cultural, gender andsocial contexts on the basis ofthe appropriate professionalismof the researcher (Morrow andRichards, 1996).D. DATA GATHERINGAND ANALYSISQuality research depends upona rational and well-consideredcombination of adequatesampling, penetrative datacollection and interpretativedata analysis techniques(Sandelowski, 2010).Researchers should be obligedto defend their sampling andanalysis strategies. Qualitativestudies generally have smallersample sizes than quantitativeresearch (Richie et al., 2003)as it is a manually intensiveapproach and large populationsizes would be unrealistic in lightof the usual budget constraintsand timeline limitations. Theconcept of ‘saturation’ isoften used in these instancesand a sample size is deemedsufficient once the collectionof further data is not giving anysignificantly new informationand the point of diminishingreturns has been reached(Mason, 2010; Richie et al.,2003). Random selection ofthe target group is particularlydifficult in hidden populationsso non-probability samplingtechniques (such as ‘snowballsampling’ or ‘respondentdriven sampling’) 3 can help insufficiently exploring conceptsaround complex phenomenasuch as CSEC, although it cannotclaim to be fully representative3.With snowball sampling, each respondent (or ‘snowball’) links the researcher up to another in the target group until thepreferred sample size is reached. In the case of respondent driven sampling (RDS) an initial participant (a ‘seed’) is asked toidentify other contacts within the target population. Each then recruits perhaps up to three more which enables the finalsample to be independent and thus more random then the initial subjects (see Goel and Salganik, 2010).4.They can include: “fear, depression, low self esteem [and] self worth, poor social skills, anger [and] hostility, inability to trust[and] build meaningful relationships in later life, blurred roles and boundaries, appearing ‘older’ (pseudomaturity), sexualizedbehaviour, guilt, shame, feeling ‘different’ from others, isolation, substance use [and] misuse, self harm (including suicide)[and] post traumatic stress disorder.”18

June, 2015of the wider target population(Morgan, 2008).As child victims of sexualexploitation are a challengingtarget group to collect datafrom due to the difficultcircumstances these childrenfind themselves in and thelinks to criminal activities andnetworks, this can lead tohesitancy to disclose or re-callbias. Using a mix of researchmethodologies and including‘ask-back’ checks can minimisebiased answering (Hassan,2005).Both quantitative andqualitative data collection andanalysis requires recording,categorising and counting ofdata, although with qualitativeanalysis ‘counting is a meansto an end, not the end itself’(Sandelowski, 2010: 338).Qualitative data analysis movesmuch more into the sphereof interpretation, rather thanquantitative statistical analysisas it is particularly helpful instudies which wish to answerthe questions who, what, howand where of a phenomena(Capaldi, 2014).CONCLUSIONWhen advocacy is based onvague, non scientific andinaccurate data it is not reliable.Whilst quality research cannotovercome all the undeniabledifficulties related to theaccessibility of data concerningan illegal activity such as CSEC,it allows relevant stakeholdersto gain reliable and detailedevidence-based knowledgeand understanding. This isessential in order to designvalid and effective policies andprogrammes to eradicate CSEC.Moreover, a scientific approachin research on CSEC is likely toincrease the political will andcommitment of relevant dutybearersto eradicating this crimeagainst children.The value of children’sparticipation is particularlysignificant in research onCSEC as their involvementgives access to qualityinformation, thus filling largegaps of knowledge aboutCSEC. Secondly, a participatoryapproach ensures that theviews and concerns of thosemost directly affected by aproblem (in this case, childrenwho are or have been sexuallyexploited) are heard. However,participatory research on CSEChas to be governed by a rightsbasedapproach. This includesfollowing strict ethical principlesand guidelines that ensurethe protection of childreninvolved in the research so asto avoid any kind of physicaland psychological harm thattheir participation may causeto them. Knowledgeable,reflective and adequatelytrained researchers can avoidprejudicial behaviour, undueinfluence and the dangers ofconflict of interest. Personsundertaking research shouldknow how to relate to thechild respondents, especiallywhen psychological assistanceand the duty of confidentialityneed to be suspended for theprotection and security of thechild. Partnerships betweenchild rights experts, civil societygroups and academics andresearch institutions can greatlyhelp in this regard.Children’s right to be properlyresearched can result inappropriately scientific,evidence-based and comparabledata from children and aboutchildren that forms the basis forsuccessful advocacy, policies andprogramme interventions.19

ISSUE : 10BIBLIOGRAPHYAkerstrom, J. and E. Brunnberg (2013). “Young people as partners in research: experiences from aninteractive research circle with adolescent girls”. Qualitative Research. Vol. 13, no. 5, 528-545.Alderson P. (1995). Listening to children, ethics and social research. London: Banardo’s.Alderson, P. (2004). “Ethics”. In Fraser S., V. Lewis, S. Ding, et al., (Eds). Doing Research with Children andYoung People. Essex: Barnardo’s.Alderson P. and V. Morrow (2004). Ethics, social research and consulting with children and young people.Berkingside: Banardo’s.Allmark, P. (2003). “The ethics of research with children”. Nurse Researcher. Vol. 10, no. 2, 7-19.Ansell, N. (2009). “Unpacking Children in Migration Research”. Children’s Geographies. Vol. 7, Issue 3.Bjerkan, L. (Ed.) (2005). A Life of One’s Own: Rehabilitation of victims of trafficking for sexual exploitation.http://lastradainternational.org/lsidocs/808%20bjerkan_l_rehabilitation.pdfBoeije, H., M. Slagt, and F. van Wesel (2013). “The Contribution of Mixed Methods Research to the Fieldof Childhood Trauma: A Narrative Review Focused on Data Integration”. Journal of Mixed MethodsResearch. Vol. 7, 347-369.Capaldi, M. (2014). The Child’s Journey in Search of Rights: A Child-Centred Study. Terre des HommesNetherlands.ECPAT International (2014). Ending Child Prostitution, Child Pornography and Trafficking for SexualPurposes: Unfinished Business. Bangkok: ECPAT International (not yet published).Edmunds, C. (2003). Ethical Considerations When Conducting Research on Children in the Worst Forms ofChild Labour. Geneva: ILO-IPEC.Ennew, J. (2008). Exploitation of children in prostitution: Thematic Paper for the World Congress III Againstthe Sexual Exploitation of Children and Adolescents. Bangkok: ECPAT International.Feingold, D. (2010). “Trafficking in numbers: the social construction of human trafficking data”. In P.A. (Eds.). Sex, drugs and body counts: the politics of numbers in global crime and conflict. CornellUniversity Press, p. 14.Goel, S. and M. Salganik (2010). “Assessing respondent-driven sampling”. Proceedings of the NationalAcademy of Sciences. 107:6743-6747. http://www.princeton.edu/~mjs3/rds.shtmlHarcourt, D. and A. Quennerstedt (2014). “Ethical Guardrails When Children Participate in Research: Riskand Practice in Sweden and Australia”. SAGE Open 2014 (4).Hassan, E. (2005). “Recall Bias can be a Threat to Retrospective and Prospective Research Designs”. TheInternet Journal of Epidemiology. Vol. 3, no. 2.Huijsmans, R. and S. Baker (2012). “Child trafficking: ‘worst form’ of child labour or worst approach toyoung migrants?”. Development and Change. Vol. 43, no. 4, 919-946.Keele (n.d.). “Quantitative versus Qualitative Research, or Both?”. Chapter 3. http://samples.jbpub.com/9780763780586/80586_CH03_Keele.pdf20

June, 2015Kovačević, I. and V. Mirović (2007). “Children Speak Out: Trafficking Risk and Resilience in SoutheastEurope: Montenegro”. Save the Children in Montenegro Child Trafficking Response ProgrammeSoutheast Europe. Montenegro: Podgorica.Knowing Children (2009). The right to be properly researched: How to do rights-based, scientific researchwith children. Bangkok: Black on White Publications.Laws, S. and G. Mann (2004). So You Want to Involve Children in Research? A toolkit supporting children’smeaningful and ethical participation in research relating to violence against children. Stockholm,Sweden: Save the Children Sweden.Mason, M. (2010). “Sample Size and Saturation in PhD Studies Using Qualitative Interviews”. Forum:Qualitative Social Research. Vol. 11, no. 3, Art. 8. http://www.qualitative-research.net/index.php/fqs/article/view/1428/3027Morgan, D. (2008). The SAGE Encyclopedia of Qualitative Research Methods. SAGE Publications, Inc.pp. 816–817.Morrow V. and M. Richards (1996). “The ethics of social research with children: an overview”. Childrenand Society. Vol. 10, 90-105.NGO Group for the CRC (2005). Semantics or Substance? www.ecpat.net/sites/default/files/Semantics%20or%20Substnce.pdfRitchie, J., J. Lewis and G. Elam (2003). “Designing and selecting samples”. In Jane Ritchie and Jane Lewis(Eds.). Qualitative research practice. A guide for social science students and researchers. ThousandOaks, CA: Sage. pp. 77-108.Sandelowski, M. (2000). “Focus on Research Methods: Whatever Happened to Qualitative Description?”.Research in Nursing and Health. Vol. 23, 334-340.SIREN (2011). The state of counter-trafficking research: researcher, programmer and donor perspectives.Bangkok, Thailand: United Nations Inter-Agency Project on Human Trafficking: Phase III.UNICEF (2008). Child Trafficking in Europe: A broad vision to put children first. http://www.unicef-irc-org/publications/pdf/ct_in_europe_full.pdfWeitzer, R. (2014). “New directions in research on human trafficking”. The ANNALS of the AmericanAcademy of Political and Social Science. Vol. 653, no. 6.21

ISSUE : 10ETHICAL ISSUES OF RESEARCHINGSEXUAL EXPLOITATION OF CHILDRENWITH VICTIMS, SURVIVORS ANDTHOSE AT RISKby Rebecca H. Rittenhouse and Elisa FeliciniINTRODUCTIONAlthough the commercial sexual exploitation of children (CSEC) has existed throughout history,it is only in the last two decades that the phenomenon has been brought to the attention ofthe international community. As the focus on CSEC grows and increased measures are taken toaddress it, there is a tremendous need for reliable, evidence-based data to support advocacy for22

June, 2015improved protection for sexually exploited and at-risk children. 1 There is therefore a vital need toconduct well-designed and varied (qualitative and quantitative, longitudinal and cohort), peerreviewedresearch on the issue. Research on the sexual exploitation of children should also bechild-centred, which ensures that the opinions, perspectives and voices of children are respectedand taken into account, including those at risk of exploitation as well as victims and survivors ofsexual exploitation. 2The participation of children in research, whether it is as a researcher or respondent, can addto the improvement of research methodologies and data collection, which can lead to moreaccurate and useful results. In some cases, participation may allow a survivor to step out of“victimization, passivity and silence,” and for those children at risk of exploitation, it has beenasserted that “a participative approach helps overcome fear and build skills to resist exploitation”(Laws and Mann, 2004).These benefits of participation may be particularly true for victims and survivors of CSEC who,due to the stigmatisation, fear of retribution and “culture of silence” that many times accompanysexual exploitation, often are not able to express or share their experiences or opinions. It hasbeen asserted that “[r]esearching with hidden populations, valuing children’s expertise andunderstanding children’s lived experience through...sensitive research has the potential to leadto positive outcomes in research and in children’s lives” (Morris et al., 2012). Additionally, ithas been suggested that excluding certain groups from research, such as abused children, mayviolate the principles of justice (Becker-Blease and Freyd, 2006).Research that involves children does raise specific ethical issues due to its sometimes sensitivenature as research questions, especially those focused on violence, abuse and exploitation,can have a harmful effect on what may be an already vulnerable child. Research that focuseson “sensitive topics” 3 has been recognised as “having the potential for creating or enhancing(existing) vulnerability among research participants” (Aldridge, 2012). These risks to a child’sphysical, emotional and psychological health should be assessed at the beginning of the researchprocess and should be taken into consideration when evaluating whether children shouldbe involved in research. While participatory research with children can be beneficial to bothresearchers and children, it is important that children are not included in research simply for thesake of including them or because researchers are curious about a certain issue, but becausethe participation will add to the quality and effectiveness of the research to improve protectioninterventions for children (Schenk and Williamson, 2005).1.ECPAT International defines CSEC as: “sexual abuse by the adult and remuneration in cash or kind to the child or a third personor persons.” The primary manifestations of CSEC include: child prostitution, child pornography/child sexual abuse materials,trafficking of children for sexual purposes and the sexual exploitation of children in travel and tourism (ECPAT International,2008). In this article, children are defined in accordance with Article 1 of the UN Convention on the Rights of the Child, as“every human being below the age of 18 years unless under the law applicable to the child, majority is attained earlier”(United Nations, 1989).2.For the purposes of this article, a victim or survivor is a child who has experienced sexual exploitation. The terms “victim” and“survivor” can be used interchangeably, although “victim” is generally preferred in the legal and medical sectors, and “survivor”in the psychological and social support sectors (IRC, 2012). In the context of this article, “survivor” is used to identifythose children who have fully recovered from the trauma of their exploitation.3.It has been recognised that “[t]here are greater barriers to participation when the research topic is sensitive, although thereis a lack of consensus between researchers, gatekeepers, parents and children as to what constitutes a sensitive topic.” Onedefinition of “sensitive topics” that has been adhered to in research studies is “those that either seem threatening, or containan element of risk in some way...[and] include areas which are private, stressful or sacred, or potentially expose stigmatisingor incriminating information” (Powell and Smith, 2009).23

ISSUE : 10These ethical considerations are particularly vital when conducting research on CSEC with children,which poses additional, specific ethical risks due to the sensitive subject matter and/or the child’sindividual experience as a victim, survivor or at risk individual. The short- and long-term effectsof sexual abuse and exploitation of children are extensive and complex 4 (Delaney and Cotterill,2005). If researchers ignore the ethical dimensions of involving children in research on CSEC, notonly will the participation not produce beneficial results, it may also harm the children involved.In recent years, there has been an increase in discussion and literature revolving around theethical considerations of the participation of children in research (Gorin et al., 2008; Powell et al.,2012; Morrow, 2013) as well as in social research that includes children’s opinions and experiences(Morrow, 2008). However, it has been observed that the discussions and participation in researchneed to further analyse “specific groups of vulnerable children” (Eriksson and Näsman, 2012). Thepurpose of this article is to examine some of the ethical issues which are raised when conductingresearch on the sexual exploitation of children with victims, survivors and those at risk as well asto highlight the key and sometimes unique ethical challenges this research poses. This article willhighlight concepts from the general ethical frameworks/guidelines for conducting research withchildren that have already been identified by others in the field such as do no harm, informedconsent, right to confidentiality and power imbalances, but will focus on research with childvictims, survivors and those at risk of sexual exploitation within that context.IS THE INVOLVEMENTOF CHILDREN THEBEST OPTION? 5Although it may have a positiveimpact on the research andthe children involved (see Lawsand Mann, 2004; Mishna et al.,2004; Schenk and Williamson,2005), involving children asrespondents in research on CSECis not always the best optionfrom an ethical point of view.In part, this is due to the factthat there is limited researchon the potential impact thatresearch procedures, such assensitive questions, may haveon child participants (Runyan,2000; Ybarra et al., 2009). Theprocess of holding consultationsor conducting research withchildren is very challenging andthere are serious considerationsto be taken into account fromthe outset in order to assessif children and young peopleshould be involved in a specificresearch project. When indoubt, the best interests of thechild should be the primaryconsideration for all researchersinvolved in research withchildren.In order to prevent harmful orunprofessional research, riskbenefitassessments need tobe made at different levels:by the researchers; by ethics,funding and scientific reviewcommittees and advisers; andby the people who are askedto take part in the research(children and young people)and their care givers (Aldersonand Morrow, 2004). In short,the research team has tofirst assess, on one hand, thepotential benefits/added valueand, on the other hand, the4.They can include: “fear, depression, low self esteem [and] self worth, poor social skills, anger [and] hostility, inability to trust[and] build meaningful relationships in later life, blurred roles and boundaries, appearing ‘older’ (pseudomaturity), sexualizedbehaviour, guilt, shame, feeling ‘different’ from others, isolation, substance use [and] misuse, self harm (including suicide)[and] post traumatic stress disorder” (Delaney and Cotterill, 2005).5.There are various forms that a “participatory approach” to research involving children can take (children as research respondents,children as data collectors/interviewers, children as observers, children involved in the design and/or analysis of theresearch, etc.). While the ethical considerations addressed in this article can apply to most, if not all, of these roles, for thepurposes of this article, “participation” refers primarily to children as research respondents.24

June, 2015risks/negative implicationsat different levels resultingfrom children’s participationin the specific research orconsultation. Second, theproject team should weigh therisks of the research againstthe benefits it hopes to achievein order to decide if directlyinvolving children is the bestoption in each specific situationor stage of the research. Inother words, the best and mostethical research strategy allowsthe collection of useful andaccurate data that responds tothe research questions whileensuring the highest protectionfor the children involved, thelatter being the priority.In the planning phase of theresearch, it is important forthe project team to explorethe motivations for seeking toinvolve child victims, survivorsor those at risk in the project,the risks the children and theproject team face by embarkingon the project; and the abilitiesand responsibilities towardsthe children involved. As partof its Ethical Research InvolvingChildren (ERIC) Project, UNICEFprovides a list of “Questionsto Guide Ethical ResearchInvolving Children.” In theplanning and preparationstage, these questions askwhat new knowledge childrenwill contribute, what will bethe likely benefits for theindividual child participating inthe research and how children’ssafety will be ensured during theresearch process (Graham et al.,2013).It has also been assertedby some that the researchitself should aim to produce“nontrivial” findings; forexample, “results that provideanswers to questions importantto the welfare of children – orthat hold substantial promiseof benefit to children” (Kingand Churchill, 2000). Theseconsiderations and questionsprior to the start of the researchare vital when seeking toinclude victims, survivors and/or children at risk of sexualexploitation to ensure not onlythat they are not harmed, butthat child victims and survivorsare not re-victimised by theirparticipation in the research.PROTECTION OFCHILD PARTICIPANTSFROM HARMThe protection of the physical,psychological and social wellbeingof children and youngpeople who are involved inresearch is paramount andmust be the main concernof researchers who have theresponsibility to ensure thatno harm will affect children asa result of their participationin the research, as well as toguarantee that children’s rightsand interests are protectedthroughout the researchprocess. This means thatpotential risks and the impact ofparticipation on the child mustbe assessed before the researchprocess begins (see Alderson,1995; Morrow and Richards,1996; King and Churchill, 2000;Morris et al., 2012). Researchersmust evaluate to what extentthe participation of a child inthe research will increase his orher exposure to violence andother risks. Specific protectionmeasures, provisions andsupport will need to be includedin the research design wheninvolving vulnerable childrenin the research such as thoseat risk or victims and survivorsof CSEC (see Gorin et al., 2008;Mudaly and Goddard, 2009;Graham et al., 2014).The commercial sexualexploitation of childrenmay include many differentforms of physical, sexual andpsychological abuse. Regardlessof their stage in recovery,involving children and youngpeople in research on sensitivetopics, such as CSEC, meanspotentially exposing themto “[p]hysical retribution,punishment or harm fromothers for participating inresearch activities” (Grahamet al., 2013; see Alderson andMorrow, 2011).Beyond potential physical harm,children involved in research onCSEC have to deal with distressthat may be caused by talkingabout their past traumaticharm may occur“through therevealing ofstigmatisinginformation abouta child within thecommunity asa consequenceof researchparticipation. Thismay occur, forexample, whenchildren...areexploited. Theremay be a risk ofdetrimentallyaffecting a child’sposition in their socialsphere or network”25

ISSUE : 10experiences of violence, abuseand exploitation. Althoughchildren may be willing totalk about their experiences,recalling the details maycause them pain and furtheremotional trauma (Twum-Danso, 2004).In addition to the psychologicaland physical harm, theparticipation of children inresearch may lead to negativeconsequences at a social level.According to UNICEF, harm mayoccur “through the revealing ofstigmatising information abouta child within the communityas a consequence of researchparticipation. This may occur,for example, when children...areexploited. There may be a risk ofdetrimentally affecting a child’sposition in their social sphere ornetwork” (Graham et al., 2013).Possible risks at a social levelcan be identified and avoidedby involving children in the riskassessment phase. However,children must be aware of suchrisks; therefore, before theresearch starts, the project teammust clearly illustrate to childrenthe aims and the outcomes(both expected and unexpected)of the research and thus obtainchildren’s “informed consent”(as discussed in detail furtherbelow).Given all these risks, although itis not always easy to determinein advance what might beemotionally or psychologicallyhurtful to someone - especiallya child - the project team mustexplore from the outset of theresearch project all the potentialphysical and psychologicalrisks to which children may beexposed during the researchand make it a priority whendesigning their researchmethodologies and tools toavoid “harm, re-traumatisation,and emotional draining of therespondents” (Bjerkan, 2005).Researchers interviewing childvictims, survivors and thoseat risk should be trained toconstruct supportive questionsand to recognise signs of distressin the children throughout theinterview and how to handlethese situations. This requires“recognition that responsibilitytowards the child is moreimportant than responsibilityfor the continuation of theresearch.” One way researcherscan accomplish this is to limitdiscussions with children tothose areas with which they feelcomfortable or are trained todeal (Mann and Tolfree, 2003).Interviewers and supervisorsshould be trained before anyresearch begins, “to ensurethat they know how to putchildren at ease and respondto needs that they might revealduring the interviews, such aspsychosocial support” (Schenkand Williamson, 2005).Some studies with childrenhave dealt with this issue byconcentrating on using researchquestions that are focusedon the present and futuresituations of the children, ratherthan the past. The reasoningbehind this strategy is that, “byfocusing on coping strategiesin the present rather thanon traumas of the past, therespondents would be able tohighlight – and perhaps alsoengage in an empoweringprocess of realising – their ownstrengths” (Bjerkan, 2005).Due to the very realisticpossibility that questionsposed to a victim or survivor,even if not directly relatedto the trauma/exploitationhe or she experienced, couldtrigger a negative emotionalresponse, researchers must beflexible enough to carry out anunstructured interview. 6CONTEXT OF THEVICTIM/SURVIVOREXPERIENCEBroad ethical guidelines andframeworks for research withchildren 7 are useful and widelyaccepted, but cannot be applieduniversally and should allowroom for personal choices byresearchers regarding ethicalconsiderations (see Morrowand Richards, 1996; King andChurchill, 2000; Christensenand Prout, 2002; Morrow,2013; BabyLaw Okoli, 2014).The participation of childrenin research is shaped by thesocial, cultural, economicaland political contexts in whichit takes place (Morrow, 2013;Graham et al., 2014). Asone researcher put it, broadguidelines and frameworks “are6.This includes stopping the questioning that is causing distress and changing the topic of discussion to something with whichthe child is more comfortable or, in some cases, gently stopping the interview.7.Eg. UK National Children’s Bureau Guidelines for Research with Children and Young People. See: http://www.nfer.ac.uk/schools/developing-young-researchers/NCBguidelines.pdf26

June, 2015minority world constructs whichcannot be applied prescriptivelyand can easily be misconstruedand misunderstood inmajority world contexts asthey do not take cognisanceof other cultures’ meanings,understandings and experiencesof children” (BabyLaw Okoli,2014).The flexibility of researcherswhen conducting researchwith child victims, survivorsand those at risk of sexualexploitation is particularlyneeded, as it has been assertedthat general ethical guidelines“frequently lack...the capacityto address the complexityof working with vulnerablepopulations” (Pittaway et al.,2010). It can also be useful whendesigning a mixed-methodologyapproach to research withchildren which can help childrenexpress themselves morefreely and make them morecomfortable (see Morrow andRichards, 1996; Morrow, 2008;Dockett et al., 2009; Pinter andZandian, 2012), especially thosethat are most vulnerable suchas child victims and survivors ofsexual exploitation (see Pinterand Zandian, 2012).SELECTION OF CHILDPARTICIPANTS ANDRESEARCHERSAn ethical approach mustbe adopted during theidentification and selectionof children at risk, victims orsurvivors as participants inresearch. In order to minimisethe harms that can result fromparticipation in research, theage and maturity of the childas well as the stage a victimor survivor has reached in therecovery process should bedecisive factors. It is vital thatchild victims or survivors whoparticipate in the research havereached an advanced stage inthe recovery process, meaningthat they have overcome thesuffering they have enduredand feel able to talk about theirexperiences without regressingin the healing process. Thisrequirement plays into theirability to consent as well asgreatly minimises the risk ofre-victimisation. In this waythey will be able to positivelygain from their participationand further develop theirself-esteem, self-worth andconfidence.When conducting research withchild victims, survivors or thoseat risk of sexual exploitation,specialised interviewersshould be recruited with theskills and expertise to conductresearch on sensitive issuesand with vulnerable children.These interviewers shouldhave experience workingwith children and be trainedto respond to the victims’/survivors’ specific needs; if theappropriate interviewers arenot available then the researchshould not be carried out(Schenk and Williamson, 2005).In research involving childrenand young people, researchersshould be recruited not juston the basis of professionalskills, but also on their abilityto relate to and work withchildren – especially vulnerablechildren such as child victims,survivors or those at risk ofsexual exploitation. In particular,researchers should have theskills that enable them toestablish a relationship based ontrust with the child participantsand allow the children to feelat ease in talking about theirexperiences.An agreement on a Codeof Conduct, in whichthe acceptable role andresponsibility of researcherswhen interacting with childrenis determined, should bedrafted before the researchbegins (Edmonds, 2003). Aswith the ethical guidelines,the Code of Conduct does notprovide researchers with allthe answers they need, but itcontains the basic principles andapproaches that should informthe researchers’ decisions.INFORMED CONSENTEach victim, survivor, or atriskchild’s story is unique andtheir personal capacities andwillingness to speak of their(or others’) experiences ofexploitation differs. For thisreason their participation shouldalways be based on the principleof “informed consent.” 8“Informed consent” consistsof four main features: (1) anexplicit act (for example, verbalor written agreement); (2) canonly be given if the participantsare informed about and have an8.Informed consent has been defined as: Agreement for voluntary participation of a participant in research, based on the individualfully understanding the goals, methods, benefits and risks of the study. Informed consent is given on the understandingthat the participant can change his or her mind about taking part in the research at any time (RWG-CL, 2002).27

ISSUE : 10understanding of the research;(3) must be given voluntarilywithout coercion; and (4) mustbe renegotiable so that childrenmay withdraw at any stage ofthe research process (Gallagher,2009 cited in Graham et al.,2013).The principal challenge in thisarea of research with childrenis how to determine if real“informed consent” by a child isachieved. It has been assertedthat consent is “informed” only“if participants understand thenature of the research and theuses to which it will be put”(Mann and Tolfree, 2003). Thisunderstanding can include,among other things, the natureof the study and how it will beconducted, the different stagesof the research process, theresearchers’ expectations of thechild and his or her role in theprocess, how the findings will beshared and how the results willbe used (Dockett et al., 2009).It is important to keep in mindthat although information ona project is provided, it is notpossible to guarantee that achild has understood it, evenwhen clear and simple languageis used.Although there is no way ofguaranteeing that children haveunderstood the information(it cannot be assumed thatthey will tell you if they havenot understood), researchersmust do their best at thebeginning of the researchprocess to provide full, clearand honest information aboutthe research project and whatthe participation of childrenwill involve in simple, clear andconcise language (Edmonds,2003). This is especiallyimportant when conductingresearch with vulnerablechildren such as CSEC victimsand survivors, as they may notalways understand the longterm affects of participation ontheir lives (see Pittaway, 2010).Biological age has long beenused as a factor to determinechildren’s ability to understandthe information provided andthus give their consent toparticipate in a project (Morrowand Richards, 1996). However,it can be argued that it is lessa question of age than ofmaturity, as children mature atdifferent rates, dependant onsuch variables as their socialand cultural environments,backgrounds, experiences andgender. Gillick-competence 9highlights that a person’sage does not necessarilydetermine competence andstates that “a competentchild is one who ‘achieves asufficient understanding andintelligence to enable him orher to understand fully whatis proposed’” (Morrow andRichards, 1996).A child’s ability to consentto participate in researchdepends on the context of theresearch and each participant’sexperiences and should bedetermined on an individualbasis (Morrow and Richards,1996; King and Churchill, 2000;Graham et al., 2013). Thus,parental consent canbe waived when it“is not a reasonablerequirement underthe circumstancesand when specialsafeguards are inplace to protectthe child subject’srights and interests,for example,when parents areunavailable or whenthe request forpermission might putthe child at risk ofharm”rather than using age as ameasuring stick for obtainingchildren’s consent, it is advisablefor researchers to focus ontheir maturity (or lack thereof),and this can only be donewithin the context of building arelationship with them.In general, in order to involvechildren in research it is usuallynecessary to seek the consentof parents or care givers. Whileadult consent is important,consent of the child should beobtained as well. (Twum-Danso,2004; Laws and Mann, 2004).There is also precedence toshow that parental consentcan be waived when it “is not areasonable requirement underthe circumstances and whenspecial safeguards are in place9.Gillick competence is a term originating from a legal case in the United Kingdom and refers to “the assessment that doctors[] make in regards to whether a child under 16 has the capacity to consent to treatment without parental or guardian consent.”For further information, see: http://www.ministryofethics.co.uk/index.php?p=7&q=2 and http://www.nspcc.org.uk/preventing-abuse/child-protection-system/legal-definition-child-rights-law/gillick-competency-fraser-guidelines/.28

June, 2015to protect the child subject’srights and interests, for example,when parents are unavailable orwhen the request for permissionmight put the child at risk ofharm” (King and Churchill,2000). 10 Even when conductingresearch on sensitive issues suchas abuse and violence, childrencan be capable of consentingif deemed mature enough (USDepartment of Health andHuman Services, 1996; Kingand Churchill, 2000; Morris etal., 2012). It has been assertedthat, regardless of age, childrenshould be permitted to consentto participation without theconsent of their parents if theycan demonstrate understandingof the research and their rights(Coyne, 2010).Obtaining support for researchon CSEC with children fromcommunities is not an easytask as parents, employers andothers may show resistanceand unwillingness to allowtheir children to participate inresearch addressing the issue ofcommercial sexual exploitationof children. This may be due to:(1) The sensitive nature of thetopic, which is still taboo andnot recognised in many parts ofthe world as a problem; (2) Thestigmatisation that accompaniesvictims and survivors of CSEC;(3) Concern about how theresearch/consultation willaffect them (both childrenand the community as awhole); (4) Concern that theproject may interfere with theincome-generating activitiesin which the children areinvolved; (5) Lack of trust ofmembers of the research team(this is especially the case ifthe team are strangers withno prior connection to thecommunity). In order to addressthis resistance, alliances andpartnerships can be made withlocal organisations that arecommitted to the same issues.However, “the need for childrento be given a voice must notbe forgotten in responding tothe views of adults” from theaffected communities (Laws andMann, 2004).Regardless of the consent given,child participants must beaware that during the researchthey have the right to changetheir minds, to withdraw or toanswer some questions but notothers (King and Churchill, 2000;Alderson and Morrow, 2004;Mishna et al., 2004). This is inpart because the implicationsof participation may becomeclear to a child only during theprocess of the research. Theresearcher should make clear tothe children that saying “no” isacceptable and that their wisheswill be respected without anynegative consequences (Lawsand Mann, 2004).Consent is an ongoing processand should be considered duringeach phase of the researchproject (Morrow, 2008).ADDRESSING POWERIMBALANCESIt has been recognised that oneof the biggest ethical challengesfor researchers workingwith children is the inherentimbalance of power between anadult researcher and a child whois participating as a respondent(Morrow and Richards, 1996).Although in research on CSECchildren are encouraged toparticipate, it is common thatwhen adults and childrenwork together on a project,adults tend to hold the powerno matter how participatorythe environment. This powerdynamic needs to be counterbalancedin the research onsexual exploitation of children inorder to avoid the manipulationof children’s contribution andallow children to effectivelyparticipate and voice theirpoints of view.When conducting researchwith child victims, survivorsand those at risk of abuse andexploitation, it is recommendedto take a child-centred approach(see Mudaly and Goddard,2006; Överlien and Hydén,2009; Morris et al., 2012).Among other things, this canaid in addressing the powerimbalance between researchersand participants by includingchildren in decision-makingand making them collaboratorsthroughout the research process(Hart, 1992; Eriksson, 2010;Morris et al., 2012).Involving children in the designand management of theresearch project can includethe participation of childrenin co-determining the data orusing a variety of data collectionmethods to encourage childrento participate and speak morefreely (Morrow and Richards,1996). Their involvement atthis level of the research can10.Other possible exceptions to parental consent include when children have the status of “emancipated” or “mature” minor(See King and Churchill, 2000).29

ISSUE : 10add to their understandingof the research process andenable them to feel a senseof ownership of the projectas well as help to developtheir confidence and a greatersense of self-worth. As aresult, this may “facilitate thecommunication and interactionbetween children and adults”(Twum-Danso, 2004).Another way to addressthe power imbalance is forresearchers to build trust withthe children. Researchersinvolve victims and survivorsin studies on CSEC in order tocollect first-hand informationand their contributions canbe invaluable. The risk is thatresearchers may treat thesechildren as mere providers ofinformation, forgetting thatthey are particularly vulnerabledue to their young age andthe exploitation they haveexperienced. This often meansthat researchers organise asingle meeting or consultationwith children, ignoring whothey are as individuals andnot involving them in followup activities and feedbackprocesses. This approach notonly fails to respect the dignity,life history and sensibility of thechildren, which can contributeto the power imbalance, but isalso likely to result in low qualityresearch due to the possiblyunreliable information thechildren will provide.This power imbalance canalso affect children’s informedconsent. As previouslymentioned, children have theright to withdraw from researchat any time; however, due tothe differing social and powerdynamics of each context,children may not feel able todo this. For example, it maybe difficult for children tovoice their discomfort or stoptheir participation due to aperceived anger or disapprovalby the adult researcher. Toaddress this, researchers candiscuss and practice variouswithdrawal strategies withchildren. It is equally importantfor researchers to “attend tochildren’s visual, verbal andnon-verbal cues to monitorunspoken expressions of uneaseor dissent” (Graham et al.,2013).In the case of research withchild victims and survivors ofsexual exploitation, the powerimbalance can be addressedto a certain extent, as in manysituations project teams do nothave direct experience with theissue. This is an opportunityfor researchers to presentthemselves not as experts, butas novices who want to learnfrom the children. In this waythe power dynamics can bereversed to some extent, “as theresearcher becomes the studentand the child, the expert”(Twum-Danso, 2004).PSYCHOSOCIALSUPPORTThrough their involvementin research, children andyoung people are likely to facepsychological and emotionaldistress, depending on thesubject matter and their ownindividual experiences. This isespecially true when involvingvictims and survivors of sexualabuse and exploitation inresearch (see King and Churchill,2000). While CSEC victimsshould be included in researchonly when they have reached anadvanced stage of the recoveryprocess or transitioned intosurvivors, re-victimisation isstill possible. Disclosure of ongoingabuse is also a possibilitywhen conducting such research(Peled, 2001; Mudaly andGoddard, 2009; Morris et al.,2012).It is important not to confuseresearch with therapy andmembers of the research teamshould not assume the roleof therapist if not trained. Itis crucial that researchers/facilitators know what type ofquestions can be asked andwhich should be omitted, inpart based on the relationshipthey have built with the child.A comprehensive researchplan should be developed toavoid re-traumatisation of childvictims and survivors of sexualexploitation. This researchplan can include a “disclosureprotocol” which describes thesteps researchers should takeif the child reports abuse by aparent or other known person,or reports engaging in “risky,self-harming behaviours.” Thisplan can also include a list ofavailable support services anda follow-up process for childrenonce the research is completedsuch as debriefing sessions,counselling or referrals (Morriset al., 2012).The psychosocial supportprovided needs to be structuredand local organisations workingon such issues can be invitedto provide such support. Ifthere is no local NGO workingon the commercial sexualexploitation of children orsexual abuse with the capacityto provide such support, itmay be worth considering30

June, 2015partnering with organisationsworking on other sensitiveareas relating to children whorequire psychosocial care, likeorganisations working withchildren with HIV/AIDS orsuffering from domestic violence(Laws and Mann, 2004). Ifappropriate psychosocialsupport cannot be provided,the project team will need toseriously consider whether toinvolve children in the project.RIGHT TOPRIVACY ANDCONFIDENTIALITYWITHIN A CHILDPROTECTIONFRAMEWORKIt is vital when conductingresearch with children thattheir identities are protectedand that the information theyprovide is kept confidential.Child victims and survivors ofabuse and exploitation may beparticularly concerned aboutanonymity when involvedin the research process dueto the fear of retribution byabusers for speaking abouttheir experiences or the stigmathat accompanies CSEC-relatedoffences in many parts of theworld. It is therefore vital that,when conducting researchwith child victims and survivorsof sexual exploitation, theresearch team creates a strategyof how to deal with issues ofconfidentiality. This strategyshould identify the specificsituations in which informationshared by a child would bedisclosed in order to protect thechild’s – or another’s – safety(Shaw et al., 2011). Accordingto UNICEF, “[t]he lack ofconsistency across internationalcontexts, within countries andacross ethical review boardsunderscores the importanceof researchers consideringthe issue of reporting prior tostarting data collection andcreating a plan or protocol tofollow if required” (Graham etal., 2013).When a research project isfocused on a specific targetgroup (eg. research on sexualabuse of children working onthe streets in a limited area orresearch on HIV/AIDS prevalenceamong children involved incommercial sexual exploitation),it can be difficult to protect theidentity of children participatingin the research. One strategyused to address this concernis to increase the samplesize to include a wider targetpopulation with a variety ofchildren and relevant questionscan be asked to specificsegments of the target groups(Edmonds, 2003).Even when these additionalprotective measures are taken,project teams need to givecareful thought to what theymean when they tell a childor parent/guardian that theirparticipation in the researchor consultation will take placeon a confidential basis. Someorganisations make it a policythat complete confidentialityshould not be assured, either toparents or children themselves,as during the project a childmay take the opportunityto disclose that he or she orothers are at risk. It is widelyacknowledged that there mustbe limits to confidentiality inresearch when child protectionin certain contexts is a concern(Wilkinson, 2000; Peled, 2001;Shaw et al., 2011). Specifically, achild’s confidentiality cannot beprotected if he or she discloses,during the course of theresearch, information that putsthe child in harms way (Jamesand Christensen, 2008; Morris etal., 2012).Breaching confidentiality canbe controversial and opinionsand practices vary in regardto how to handle such asituation (Cashmore, 2006).For example, the child’s right toconfidentiality may clash withthe ethical responsibility ofthe researcher to ensure thathe or she is protected fromharm (Graham et al., 2013).Researchers also need to beaware of requirements formandatory reporting as“[t]he decision to reportconcerns or knowledgeregarding harm or potentialharm to children may be a legalone as well as an ethical one”(Fisher, 1994 cited in Graham etal., 2013).Researchers should discuss theissues of confidentiality withchild victims, survivors andthose at risk at the beginningof the research process toascertain what concerns them,which may vary on the basis ofeach child, the situation, as wellas the social and cultural contexthe/she comes from (Twum-Danso, 2004). They shouldinform children of the limits ofthe confidentiality agreementfrom the outset of the researchand again during the interviewwhen it appears that a childis about to disclose sensitiveinformation.31

ISSUE : 10SHARING THERESEARCHOUTCOMESThe participation andcontribution of child victims,survivors and those at risk inresearch on sexual exploitationis of paramount importancebefore and during the study;however, once the data iscollected, they often get leftbehind and forgotten. In fact,very often, children and theircommunities, especially thosewho are isolated and hardto reach, do not receive anyfeedback on the outcomesof a project in which theyparticipated. For childparticipants, it is critical that theoutcome of the child-centredresearch is shared (Edmonds,2003), especially those whosevoices have traditionally beensilenced such as child victims ofcommercial sexual exploitation.Therefore, once the project hasbeen completed, it is importantthat researchers not only sharethe final results of the researchwith the children, but give themthe opportunity to contributetheir input before the finaldissemination so that they canvalidate the information andcorrect any misinterpretations(Twum-Danso, 2004). Sharingresults with children can bechallenging due to languagebarriers, illiteracy and lackof accessibility. However,the benefits for the childreninvolved and the researchresults are so important that “adetermined effort must be madeto include such an activity withinthe overall framework of theresearch process” (Edmonds,2003).CONCLUSIONSAlthough there are significantbenefits and positiveimplications to involving childvictims, survivors and those atrisk to sexual exploitation inresearch, it is clear that suchan approach is not immune toethical concerns that may ariseduring all stages of the researchprocess. The participation ofchildren should be generallyencouraged in order to respecttheir right to voice their pointof view on a phenomenon thataffects many of them. It shouldbe emphasised, however, thatthis approach is not alwaysthe most effective option or inthe best interests of the child,especially when dealing withvulnerable children such as childvictims, survivors and those atrisk of sexual exploitation.Child victims and survivors ofsexual exploitation can gainextraordinary benefits fromsharing their views on mattersthat affect them and feelingas though, through theirparticipation in research, theyare helping other vulnerablechildren. The short- andlong-term effects of sexualexploitation on each child aredifferent, however, and theexperience of each victim,survivor or at-risk child is basedon his or her own cultural, socialand economic background,resilience and support system.For this reason, it is essentialthat programme managersand researchers take intoconsideration all the potentialpros and cons of involvingchildren in studies on thesexual exploitation of children.Involving victims, survivors andat-risk children in a study onsexual exploitation presentspotential invaluable benefits,not only for the researchitself but also for the childrenwho participate. However,the project team cannotignore the risks and ethicalconcerns that adopting such anapproach may entail. From thepre-planning stages throughto the dissemination of theresearch, these concerns andlimitations should be carefullyweighed against the bestinterests of the child, whichshould be the highest priorityfor all participatory research,especially research focusedon the commercial sexualexploitation of children.32

June, 2015BIBLIOGRAPHYAlderson P. (1995). Listening to children: children, ethics and social research. Barkingside, United Kingdom:Barnardos.Alderson P. and V. Morrow (2004). Ethics, social research and consulting with children and young people.Ilford, United Kingdom: Barnardo’s.Alderson, P. and V. Morrow (2011). The ethics of research with children and young people. London: SagePublications.Aldridge, J. (2012). “Working with vulnerable groups in social research: dilemmas by default and design”.Qualitative Research. Vol. 14, no. 1 (2012): 112-130.BabyLaw Okoli, R.C. (2014). “Ethical Issues and dilemmas in doing research with itinerant street vendingchildren and young people: Experiences from Nigeria”. Qualitative Social Work. pp.1-16.Becker-Blease, K.A. and J.J. Freyd (2006). “Research Participants Telling the Truth about Their Lives: TheEthics of Asking and Not Asking About Abuse”. American Psychologist. Vol. 61, no. 3 (2006): 218-226.Bjerkan, L. (2005). A Life of One’s Own: Rehabilitation of victims of trafficking for sexual exploitation.Norway: Fafo.Cashmore, J. (2006). “Ethical issues concerning consent in obtaining children’s reports on their experienceof violence”. Child Abuse & Neglect. Vol. 30, no. 9 (2006): 969-977.Christensen, P. and A. Prout (2002). “Working with Ethical Symmetry in Social Research with Children”.Childhood. Vol. 9, no. 4 (2002): 477-497.Coyne, I. (2010). “Research with children and young people: the issue of parental (proxy) consent”.Children and Society. Vol. 24 (2010): 227-237. In M.A. Powell, R.M. Fitzgerald, N. Taylor and A. Graham(2012). International literature review: ethical issues in undertaking research with children and youngpeople. Dunedin, NZ. http://epubs.scu.edu.au/cgi/viewcontent.cgi?article=1041&context=ccyp_pubs.Delaney S. and C. Cotterill (2005). The Psychosocial Rehabilitation of Children who have been CommerciallySexually Exploited: A Training Guide. Bangkok: ECPAT International.Dockett, S., J. Einarsdottir and B. Perry (2009). “Researching with children: ethical tensions”. Journal ofEarly Childhood Research. Vol. 7, no. 3 (2009): 283-298.ECPAT International (2008). Questions & Answers about the Commercial Sexual Exploitation of Children,4th Edition. Bangkok: ECPAT International. http://ecpat.net/sites/default/files/faq_eng_2008.pdfEdmonds C. N. (2003). Ethical Considerations When Conducting Research on Children in the Worst Formsof Child Labour. Geneva: ILO-IPEC.Eriksson M. (2010). “Children who’s witness’ violence as crime victims and changing family law in Sweden”.Journal of Child Custody. Vol. 7, no. 2 (2010): 93-116.Eriksson, M and E. Näsman (2012). “Interviews with Children Exposed to Violence”. Children & Society.Vol. 26 (2012): 63-73.Fisher, C. (1994). “Reporting and referring research participants: ethical challenges for investigatorsstudying children and youth”. Ethics & Behaviour. Vol. 4, no. 2 (1994): 87-95.Gallagher, M. (2009). “Ethics”. In E.K. Tisdall, J. Davis and M. Gallagher, eds. Researching with children andyoung people: Research design, method and analysis. London: Sage Publications.33

ISSUE : 10Gorin, S., C.A. Hooper, C. Dyson, and C. Cabral (2008). “Ethical challenges in conducting research with hardto reach families”. Child Abuse Review. Vol. 17 (2008): 275-287.Graham, A., M. Powell, N. Taylor, D. Anderson and R. Fitzgerald (2013). Ethical Research Involving Children.Florence: UNICEF Office of Research-Innocenti. http://www.unicef-irc.org/publications/pdf/ericcompendium-approved-digital-web.pdfGraham, A., M.A. Powell, and N. Taylor (2014). “Ethical Research Involving Children: Encouraging ReflexiveEngagement in Research with Children and Young People”. Children & Society. First published: 21August 2014. DOI: 10.1111/chso.12089.Hart, R. (1992). “The ladder of participation”. In UNICEF. Children’s Participation: From Tokenism toCitizenship. Florence: UNICEF ICDC.International Rescue Committee (IRC) (2012). Caring for Child Survivors of Sexual Abuse: Guidelines forhealth and psychosocial service providers in humanitarian settings. New York: IRC.James A. and P.M. Christensen (2008). Research with Children: Perspectives and Practices. New York:Falmer Press.King, N.M.P. and L.R. Churchill (2000). “Ethical Principles Guiding Research on Child and AdolescentSubjects.” Journal of Interpersonal Violence. Vol. 15, no. 7 (July 2000): 710-724.Laws, S. and G. Mann (2004). So you Want to Involve Children in Research? A toolkit supporting children’smeaningful and ethical participation in research relating to violence against children. Stockholm: Savethe Children Sweden. http://www.savethechildren.org.uk/sites/default/files/docs/So_you_want_to_involve_children_in_research_SC_2004_1.pdfMann G. and D. Tolfree (2003). Children’s Participation in Research: Reflections from the Care andProtection of Separated Children in Emergencies Project. Stockholm: Save the Children Sweden.Mishna, F., B.J. Antle, and C. Regehr (2004). “Tapping the Perspectives of Children: Emerging Ethical Issuesin Qualitative Research”. Qualitative Social Work. Vol. 3, no. 4 (2004): 449-468.Morris, A., K. Hegarty and C. Humphreys (2012). “Ethical and safe: Research with children about domesticviolence”. Research Ethics. Vol. 8, no. 2 (2012): 125-139.Morrow, V. (2008). “Ethical dilemmas in research with children and young people about their socialenvironments”. Children’s Geographies. Vol. 6, no.1 (February 2008): 49-61.Morrow, V. (2013). “Practical Ethics in Social Research with Children and Families in Young Lives: Alongitudinal study of childhood poverty in Ethiopia, Andhra Pradesh (India), Peru and Vietnam”.Methodological Innovations Online. Vol. 8, no. 2 (2013): 21-35.Morrow V. and M. Richards (1996). The ethics of social research with children: an overview. Children andSociety. Vol. 10 (1996): 90-105.Mudaly, N. and C.R. Goddard (2006). The Truth is Longer than a Lie – Children’s Experiences of Abuse andProfessional Interventions. London: Jessica Kingsley.Mudaly, N. and C.R. Goddard (2009). “The ethics of involving children who have been abused in childabuse research”. International Journal of Children’s Rights. Vol. 17 (2009): 261-281.Överlien C. and M. Hydén (2009). “Children’s actions when experiencing domestic violence”. Childhood.Vol. 16, no. 4 (2009): 479-496.Peled, E. (2001). “Ethically sound research on children’s exposure to domestic violence: A proposal”. In S.A.Graham-Bermann and J.L. Edleson. Domestic Violence in the Lives of Children: The Future of Research,Intervention and Social Policy. Washington, DC: American Psychological Association.34

June, 2015Pinter, A. and S. Zandian (2012). “I thought it would be tiny little one phrase that we said, in a huge big pileof papers’: children’s reflections on their involvement in participatory research”. Qualitative Research.Vol. 15, no. 2 (2012): 235-250.Pittaway, E., L. Bartolomei and R. Hugman (2010). “’Stop Stealing our Stories’: The Ethics of Research withVulnerable Groups”. Journal of Human Rights Practice. Vol. 2, no. 2 (June 2010): 229-251.Powell, M.A., R.M. Fitzgerald, N. Taylor and A. Graham (2012). International Literature Review: EthicalIssues in Undertaking Research with Children and Young People, for the Childwatch InternationalResearch Network, Southern Cross University, Centre for Children and Young People, Lismore NSW andUniversity of Otago, Centre for Research on Children and Families, Dunedin, NZ.Powell, M.A. and A.B. Smith (2009). “Children’s Participation Rights in Research”. Childhood. Vol. 16, no.1 (2009): 124-142.Regional Working Group on Child Labour in Asia (RWG-CL) (2002). Handbook for action-oriented researchon the worst forms of child labour, including trafficking in children. Bangkok.Runyan, D. (2000). “The ethical, legal and methodological implications of directly asking children aboutabuse”. Journal of Interpersonal Violence. Vol. 15, no. 7 (2000): 675-681.Schenk, K. and J. Williamson (2005). Ethical Approaches to Gathering Information from Children andAdolescents in International Settings: Guidelines and Resources. Washington, DC: Population Council.Shaw, C., L.M. Brady and C. Davey (2011). Guidelines for Research with Children and Young People. London:National Children’s Bureau.Twum-Danso, A. (2004). Involving Children and Young People in Research and Consultations: EthicalGuidelines from a CSEC perspective. ECPAT International (unpublished).United Nations (1989). Convention on the Rights of the Child.U.S. Department of Health and Human Services (1996). Protection of Human Subjects. 45 C.F.R. Part 46,Subpart D – Additional DHHS Protections for Children Involved as Subjects in Research.Wilkinson, J. (2000). Briefing Facts. Children and Participation: Research, Monitoring and Evaluation withChildren and Young People. London: Save the Children.Ybarra, M.L., J. Langhinrichsen-Rohling, J. Friend and M. Diener-West (2009). “Impact of Asking SensitiveQuestions about Violence to Children and Adolescents”. Journal of Adolescent Health. Vol. 45 (2009):499-507.LINKShttp://www2.le.ac.uk/institution/committees/research-ethics/research-concerning-children-and-youngpeople-guidelineshttp://www.unicef-irc.org/http://childethics.com/http://www.nspcc.org.uk/Inform/research/briefings/ethical-research-factsheet_wda97712.html35

ISSUE : 10AUTHOR BIOGRAPHIESMark Capaldihas worked in South and East Asia with Concern Worldwide, PACT Inc., Save the Children UK andcurrently with ECPAT International. In June 2011, he started a Doctorate programme at MahidolUniversity, Bangkok.Elisa Felicinihas a Master’s Degree in Human Rights and Peace Studies from the University of Padua in Italy. Shehas carried out field research about the trafficking of children and CSEC in Romania in 2008 andhas worked with various NGOs – in Italy as well as in other countries – committed to protectingchildren’s rights, including ECPAT International in Bangkok from June 2013 – April 2014.Jordana Dawson HayesJordana Dawson Hayes obtained a Bachelor of Science and Bachelor of Law (Hons) from theUniversity of Waikato, New Zealand. She has worked with a variety of NGOs in the areas of law,policy and evaluation. She currently works at the Fred Hollows Foundation NZ as an Evaluationand Policy Analyst.Rebecca H. Rittenhouseobtained her law degree from New England Law/Boston in the United States in 2011. She hasworked as a researcher for the Human Rights Foundation in New York City and is the Researchand Monitoring Officer for the ECPAT International Secretariat in Bangkok, Thailand.36

June, 201537

ECPAT International328/1 Phayathai RoadRatchathewi, Bangkok10400 THAILANDTel: +662 215 3388, +662 611 0972Fax: +662 215 8272Email: info@ecpat.netWebsite: www.ecpat.net