course outline (pdf)

Course title
ECPR Summer School on Methods and Techniques
Course Description Form
B3. Experimental Methods
Instructor details
Week 1
First name, last name: Wolfgang J. Luhan
Department/Unit: Institute for Macroeconomics & Bochum Lab for Experimental Economics
(RUBex)
Institution: Ruhr-University Bochum
Full postal address for ECPR correspondence : Lehrstuhl f. Makrooekonomik , Ruhr-Universitaet
Bochum, Universitaetsstrasse 150, 44801 Bochum, Germany
Phone: +49 (0) 234/32-25329
Fax: +49 (0) 234/32-05329
E-mail : wolfgang.luhan@rub.de
Week 2
First name, last name: Fabian Winter
Department/Unit: Strategic Interaction Group
Institution: Max Planck Institute of Economics
Full postal address for ECPR correspondence : Max Planck Institute of Economics, Kahlaische Str. 4,
07743 Jena, Germany
Phone: +49 (0) 3641 / 686 640
Fax: +49 (0) 3641 / 686 667
E-mail : winter@econ.mpg.de
Short Bio
Wolfgang Luhan is Assistant Professor of economics at the Ruhr-University Bochum’s Economics
department and Director of the Bochum Lab for Experimental Economics (RUBex). His research
focus lies on behavioural (macro) economics, bargaining situations and labour markets. He has been
doing experimental research since 2004 and has taught numerous classes on experimental research
methods.
Fabian Winter is Research Associate at the Max Planck Institute for Economics. He holds a PhD in
Sociology with a strong focus on Economics and the social sciences more generally. Fabian’s research
interests include the emergence and evolution of social norms, trust relations, social networks, or
signaling. He has been actively involved in experimental research since 2004 and has taught classes on
experimental methods and statistics.
Short outline
The course offers an introduction to the methods and techniques of experimental empirical research,
especially laboratory and “pen-and paper” experiments (to accommodate research in less technically
equipped settings). It pursues the (rather ambitious) agenda of providing the theoretical knowledge as
well as practical skills to design and implement laboratory experiments. Hence, the focus will be on
general methodological, design and implementation issues, rather than surveying the existing
experimental evidence in a given field of research in detail. We will follow a learning-by-doing
approach, in which students will frequently participate in experiments that demonstrate behavioural
hypothesis and approaches to test them along with interactive lectures.
At the end of this course you will have gained not only the know-how needed to develop and
implement an experimental research design in the laboratory but you have also gained the basic
statistical and software skills (STATA, Z-tree) required to gather, analyse and interpret experimental
data obtained in laboratory and the field.

Long outline
The course will be divided in two sessions per day. The daily afternoon sessions will introduce the
basic concepts, conventions and common practical problems of experimental research in the social
sciences. These lectures will be accompanied by group work and (mostly) pen-and-paper classroom
experiments to give a first look-and-feel of participating in experiments and interpreting the collected
data.
The evening sessions will take place in the computer pool and will introduce the students to the
software package zTree, designed to carry out lab experiments. The timing of the evening sessions will
depend on how fast we proceed and will cover the basics (and hopefully more) of implementing zTree
treatments. The students will learn all steps necessary to implement simple experiments hands-on in
small groups, supported by the teaching staff. We will follow the structure of the tutorial to zTree, and
students will program many treatments themselves. At the end of the two weeks course, the students
will have learned the structure and language of zTree as well as all the important features of the
program, enabling them to program their own experiments.
The lectures in week 1 will by and large cover the theoretical foundations of experiments and the
fundamental concepts of game theory underlying many experimental studies in the social sciences.
Week 2 will introduce the more practical issues of planning, implementing and running experiments.
In week 1, the course will start with group work a first experiment and an overview of the history of
experimental research in the social sciences. We will mainly focus on the development of
experimental economics, which has substantially departed from the methodological conventions of
other disciplines like e,g, social psychology. Furthermore, we stress the idea of theory-led research,
where theoretical hypotheses are tested in lab experiments. There will be no programming exercise in
zTree on the first day, but we will give an overview of the program instead.
Day 2 will discuss theoretical considerations about why one would want to run an experiment.
Experiments can be designed in order to test and to develop better theories, but also to give evidence
based advice. Experiments are particularly suited for these purposes, as they have distinct features that
enable causal inferences about the relationship between two or more variables of interest: They are
usually strong in internal validity. Yet, they do have (sometimes severe) limitations, depending on the
environment in which they are carried out and the subjects participating in them. Both, strength and
weaknesses of experimental research will be carefully discussed.
Day 3 will give a more in-depth discussion of the philosophy of experimental research more generally,
with a focus on the social sciences. We will furthermore discuss the empirical research topics covered
in the last decades and those “hot” today.
The remaining two days of week one will give an introduction into the basic concepts of expected
utility and game theory, starting with von Neumann/Morgenstern utility, the “Prisoners’ Dilemma”
game and the concepts of dominance as well as pure nash equilibria, and will continue with mixed
Nash Equilibria and how to find them in games like the “battle of the sexes”. No prior knowledge of
game theory is required here, though basic economic thinking will be helpful.
In week 2, we dive deeper into the practical issues of design implementation and analysis of
experiments.
Day 6 will start with the most fundamental assumption of experimental economics: Induced Value
Theory states that the experimentalist can translate preferences into money and thus control and
manipulate the incentives of the subject participating in the experiment. This – obviously controversial
– claim is the reason for monetary incentives in economics experiments even outside the narrow
domain of markets.
Day 7 starts with the essentials of running experiments: Terminology (what is a treatment, a session,
…), steps to take to design and run an experiment (including how to find a good research question),
do’s and don’ts of experimental research and the particular, sometimes contradicting, rules in different
academic disciplines. We will for instance discuss controversy over the rule (not) to deceive subjects
between economists and psychologists or the pros and cons of paying subjects according to their
performance.

Days 8 and 9 will be devoted to the most practical issues of experimental research. How do I design a
study? Where and how can you apply for money? In which mode should I conduct the experiment,
computerised or paper-pencil? How do I get people to come to the lab? Once they are in the lab, which
particular steps do I have to take during an experiment? These two days are planned to be less
theoretical than some of the days before and should give many useful tips from and for the
practitioner’s side. We will, for instance, discuss how to write good instructions, and how poorly
written ones can spoil an experiment.
The last day will give an introduction into non-parametric statistical methods. Other than parametric
approaches (e.g. t-test, OLS regressions), non-parametric testing makes only very limited assumptions
about the structure of the analyzed data. This is particularly helpful if the per-cell sample size is rather
small, as in most experimental studies. We will discuss the non-parametric alternative to the most
important parametric test like Mann-Whitney instead of the t-test.
5. Day-to-day schedule
- Week 1
Topic(s) Details [NB : incl. timing of lecture v/s lab or
fieldwork etc. hours]
Day 1 General Introduction/Group
work/History/z-Tree Intro
14:00 -17:00: Classroom only
Day 2 Reasons for
14:00 -15:30: Classroom
Experiments/Typology/Duhem-
Quine Thesis/Advantages &
Limitations/Validity
16:00 – 17:00: lab
Day 3 Reductionism/Methodological 14:00 -15:30: Classroom
Challenges/Major Results 16:00 – 17:00: lab
Day 4 Game Theory Intro:
14:00 -15:30: Classroom
Prisoners Dilemma
16:00 – 17:00: lab
Day 5 Game Theory Intro:
14:00 -15:30: Classroom
Battle of Sexes
16:00 – 17:00: lab
- Week 2
Day 6 Induced Valuation Theory 14:00 -15:30: Classroom
Day 7 Essentials/Terminology/Steps/
Economic vs. Psychological
Experiments/Deception/
Monetary incentives
Day 8 Design Questions
/Trials/Hypotheses/
Implementation 1: Getting
Money/Writing Instructions
Day 9 Implementation 2:
Recruiting/Programming/Prepar
ing the Session/Payoffs
16:00 – 17:00: lab
14:00 -15:30: Classroom
16:00 – 17:00: lab
14:00 -15:30: Classroom
16:00 – 17:00: lab
14:00 -15:30: Classroom
16:00 – 17:00: lab
Day 10 Non-Parametric Statistics 14:00 -15:30: Classroom
16:00 – 17:00: lab
Day 11 -
Saturday
Exam 9:00 – 12:00
Day-to-day reading list
- Week 1
Readings (please read at least the compulsory reading for the scheduled day)
Day 1 Chapter 1: Experimental Political Science in Perspective. in Kittel et al. 2012 (see
below); z-Tree Tutorial pp. 1-13
Day 2 Chapter 1.I. in Kagel and Roth 1997 (see below); z-Tree Tutorial pp. 14-25
Day 3 Chapter 1.III.F. in Kagel and Roth 1997 (see below) ; z-Tree Tutorial pp. 26-33

Day 4 Capter 1 up to 1.1.C. in Gibbons 1992 (see below) ; z-Tree Tutorial pp. 34-41
Day 5 ; z-Tree Tutorial pp. 42-52
- Week 2
Day 6 Smith, V.L. (1976): Experimental economics: induced value theory. American
Economic Review 66 (2): 274-79; Smith, V. L. (1994): Economics in the laboratory.
Journal of Economic Perspectives 8(1): 113-131. ; z-Tree Tutorial pp. 52-57
Day 7 Morton, R.B. and K. Williams (2010): Experimental Political Science and the Study
of Causality. From Nature to the Lab. Cambridge: Cambridge University Press.
Pp. 363-378
; z-Tree Tutorial pp. 57-66
Day 8 Friedman, D., and S. Sunder (1994) Experimental Methods: A Primer for
Economists. (Cambridge: Cambridge University Press).
pp.38-71 ; z-Tree Tutorial pp. 67-76
Day 9 Friedman, D., and S. Sunder (1994) Experimental Methods: A Primer for
Economists. (Cambridge: Cambridge University Press).
pp.74-82; z-Tree Tutorial pp. 77-86
Day 10 Friedman, D., and S. Sunder (1994) Experimental Methods: A Primer for
Economists. (Cambridge: Cambridge University Press).
pp.85-116; z-Tree Tutorial pp. 86-91
Requested prior knowledge
Basic economic thinking (micro-economics), deductive reasoning, and strategic thinking. Knowledge
of standard game-theoretic concepts is helpful but no precondition. Some acquaintance with statistics
and basic programming concepts will increase the benefit from the econometrics and z-Tree part,
respectively.
Software used
STATA, Zurich Toolbox for Readymade Economic Experiments – z-Tree 3.3.12
Literature
Books, Overview and Background Reading
Bardsley, N., R. Cubitt, G. Loomes, P. Moffatt, C. Starmer, and R. Sugden (2010): Experimental
Economics. Rethinking the Rules. Princeton: Princeton University Press.
Bergstrom, T.C. und J.H. Miller (2000): Experiments with Economic Principles: Microeconomics.
2. nd ed., New York: McGraw-Hill.
Camerer, C.F. (2003): Behavioral Game Theory: Experiments in Strategic Interactions. Princeton:
Princeton University Press.
Davis, D.D. und C.A. Holt (1993). Experimental Economics. Princeton: Princeton University Press.
Guala, F. (2005): The Methodology of Experimental Economics. Cambridge: Cambridge University
Press.
Holt, C. (2007): Markets, Games & Strategic Behavior. Pearson Addison Wesley
Kagel, J.H. and A.E. Roth (eds.) (1997): The Handbook of Experimental Economics. Princeton:
Princeton University Press.
Kittel, B., W. J. Luhan, and R. Morton (eds.) (2012): Experimental Political Science: Principles and
Practices. Palgrave-Macmillan. forthcoming.
Morton, R.B. and K. Williams (2010): Experimental Political Science and the Study of Causality.
From Nature to the Lab. Cambridge: Cambridge University Press.
Introduction History, Methodology
Loewenstein, G. (1999): Experimental economics from the vantage point of behavioural conomics.
Economic Journal 109 (453): F25-34.
Rabin, M. (1998). Psychology and economics. Journal of economic Literature 36 (1): 11-46.
Roth, A.E. (1988): Laboratory experimentation in economics: a methodological overview. Economic
Journal 98 (393): 974-1031.
Schram, A. (2005): Artificiality: The tension between internal and external validity in economic
experiments. Journal of Economic Methodology 12(2): 225-237.
Smith, V. L. (1973): Notes on some literature in Experimental Economics. Working Papers 21,
California Institute of Technology.

Smith, V.L. (1976): Experimental economics: induced value theory. American Economic Review 66
(2): 274-79.
Smith, V. L. (1994): Economics in the laboratory. Journal of Economic Perspectives 8(1): 113-131.
Before and after: Invitations to payoffs.
To my knowledge, there is no general “best practice guideline”. Some tips regarding implementation
can be found in
Friedman, D., and S. Sunder (1994) Experimental Methods: A Primer for Economists. (Cambridge:
Cambridge University Press).
List, J., S. Saddoff, and M. Wagner (2011): So you want to run an experiment, now what? Some
simple rules of thumb for optimal experimental design. Experimental Economics
Experimental software: z-Tree
All relevant literature will be provided, but can also be found here:
http://www.iew.uzh.ch/ztree/index.php
Non-parametric Statistics
One very popular text(yet hard to find) is:
Siegel, S. and J. Castellan (1988): Nonparametric Statistics for Behavioral Sciences. McGraw-Hill.
Introduction to game theoretic concepts
As this is not one topic but rather a whole field of research (basically one of the foundations of
modern economics) we can only recommend some introductory texts.
Rather non-technical introductions are
Dixit, A., and S. Skeath (2004): Games of Strategy.New York: Norton.
Gibbons, R. (1992): A Primer in Game Theory, Prentice Hall
A more advanced Textbook is
Myerson, R.B. (1991): Game theory: analysis of conflict. Harvard: Harvard University Press.
A textbook with applications to many domains in the political sciences is
McCarty, M. and A. Meirowitz (2007): Political Game Theory: An Introduction. Cambridge:
Cambridge University Press.
Lecture room requirement
Black board, projector, but else no special requirements.