SOS9026 – Experimental Research Designs in the Social Sciences
Course description
Course content
The interest in experimental research designs has increased rapidly in the social sciences over the last decade, partly as a result of increased awareness of the problems of causal inference without direct control of assignment to treatment. This course offers an introduction to experimental research designs tailored in the potential outcomes framework of causal inference. The course will cover field, lab and survey experiments. The logic of experimentation, the power of randomization, and its strengths and weaknesses compared to other methodologies will be discussed. Students will learn how to design randomized experiments, how to analyze the data, and how to interpret the findings. Applied examples from across the social sciences will be used to illustrate the strengths and potential weaknesses of various experimental designs. In addition, there will be lectures presenting results from experiments conducted in Norway. Students are expected to actively participate and to conduct homework between the lectures.
The course is administered by the Department of Sociology and Human Geography, but the course should be of equal interest to students and researchers in other fields of social science, including sociology, political science, economics and psychology.
Admission
This course is aimed at doctoral students, researchers and talented master students in the social sciences with an interest in experimental methods. Participants should have a good working knowledge of applied regression analysis. Prior exposure to Stata is an advantage.
Ph.D. students at the Department of Sociology and Human Geography register for the course in Studentweb. For Ph.D. students at the Department of Sociology and Human Geography, course participants obtain “method” points.
Interested master students and participants outside the Department of Sociology and Human Geography shall fill out this application form.
The application deadline is four weeks prior the course: November 8, 2015.
Teaching
The lectures will take place in PC-room 035 in the basement of Harriet Holter Building
Literature
- Alan S. Gerber, Donald P. Green, Field Experiments: Design, Analysis, and Interpretation (2012)
- Alan S. Gerber, Donald P. Green, The Effects of Canvassing, Telephone Calls, and Direct Mail on Voter Turnout: A Field Experiment (2000) read here
- Arnfinn H. Midtb?en, Discrimination of the Second Generation: Evidence from a Field Experiment in Norway (2014), read here
- Arnfinn H. Midtb?en, The context of employment discrimination: interpreting the findings of a field Experiment (2015), read here
- Henning Finseraas, Andreas Kotsadam, Does Personal Contact with Ethnic Minorities Affect Support for Welfare Dualism? Evidence From a Field Experiment (2015), read here
- Henning Finseraas, ?shild A. Johnsen, Andreas Kotsadam, Gaute Torsvik, Exposure to female colleagues breaks the glass ceiling - Evidence from a combined vignette and field Experiment (2015), read here
- Kevin Arceneaux, Alan S. Gerber and Donald Green, Sociological Methods & Research (2010), read here
Schedule
8th December:
9.00-11.30: Session 1. What are experiments? Why conduct experimental research? FEDAI: Chapter 1.
11.30-12.30: Lunch
12.30-14.30: Session 2. Experiments and Models of Potential Outcomes. FEDAI: Chapter 2
14.30-15: Coffee break
15.00-17.00: Session 3. Blocking, Clustering, and Covariate Adjustment. FEDAI: Chapter 3 and 4. In addition, read the following article, which illustrates the use of blocking.
9th December:
9.00-11.30: Session 1. Computer Practicum Analyzing Experimental Data
11.30-12.30: Lunch
12.30-14.30: Session 2. Field Experiments with One-sided Noncompliance (Failure-to-Treat). FEDAI: Chapter 5. In addition, read the following article, which we will use in class to illustrate the analysis of experiments with one-sided noncompliance.
14.30-15: Coffee break
15.00-17.00: Session 3. Computer Practicum Analyzing Experimental Data with Noncompliance
10th December:
9.00-11.30: Session 1. Complications Posed by Attrition, Spillovers, and Heterogeneous Effects. FEDAI: Chapters 7, 8
11.30-12.30: Lunch
12.30-15.00: Session 2. Computer Practicum Analyzing Spillovers and Heterogeneous Effects
14.30-15: Coffee break
15.00-17.00: Session 3. Implementing a Field Experiment and Reporting the Results. FEDAI: Chapter 13, Appendix A, and Appendix B
11th December:
9.00-11.30: Session 1 (Arnfinn H. Midtb?en). Ethnic Discrimination in the Norwegian Labour Market: A Field Experiment. Midtb?en 2014, 2015
11.30-12.30: Lunch
12.30-15.00: Session 2 (Henning Finseraas). Causal effects of exposure to ethnic minorities and female colleagues: Results from field experiments in the Norwegian Army. Finseraas et al 2015 a, b
15.00-15.15: Coffee break
15.15-16.00: Discussion
The main part of the course will be led by Donald P. Green, professor of political science at Columbia University. Professor Green has published in top journals across the social sciences on topics including voting behavior, partisanship, campaign finance, hate crime, and research methods. He is the author, together with Alan S. Gerber, of Field Experiments: Design, Analysis, and Interpretation (2012), which serves as the main text for the course.
In addition, Arnfinn H. Midtb?en (PhD Sociology) and Henning Finseraas (PhD Political Science), both researchers at Institute for Social Research (ISF), will hold lectures presenting results from experiments conducted in Norway.
Contact: Arnfinn H. Midtb?en a.h.midtboen@samfunnsforskning.no and Henning Finseraas henning.finseraas@samfunnsforskning.no
Examination
Participants obtain 6 ECTS credits by completing the course requirements, which are (1) active participation throughout the course, (2) completion of three 1 hour problem sets on the topics covered during the preceding day, and (3) composition of a 2500 word essay that describes the results of a practicum experiment that each student will design, implement, and interpret. This experiment will involve at least 30 subjects and should involve an intervention that is inexpensive to implement, poses no risks of harm to subjects, and raises no ethical concerns. For example, students may elect to conduct a test of a physical process (e.g., does water freeze faster when its starting temperature is 10C than 5C?), an intervention with non-humans (e.g., do dogs in a dog park respond differently to people of different heights?), or an in intervention with humans (e.g., are expensive wines rated more highly in blind taste-tests than inexpensive wines?). Write-ups should follow the reporting guidelines laid out in Chapter 13 of FEDAI.
The deadline for delivering the essay is 22nd Desember. Send it to katalin.godberg@sosgeo.uio.no.
Grading scale
Grades are awarded on a pass/fail scale. Read more about the grading system.