Mr Thomas Leeper CON 3.21

This course is available on the BSc in Government, BSc in Government and Economics, BSc in Government and History, BSc in Philosophy, Politics and Economics, BSc in Politics and International Relations and BSc in Politics and Philosophy. This course is not available as an outside option nor to General Course students.

This course is capped at one group. The deadline for enrolments is 12 noon on Friday 29 September 2017.

Familiarity with basic algebra required and comfort with basic statistics as covered by Research Design in Political Science (GV249) or an equivalent course in research design or introductory statistics (such as ST102, ST107, ST108, GY140, SA201) is recommended.

This course will introduce students to the use of experiments or randomised controlled trials (RCTs) in politics to evaluate policies, programmes, and theories. The course will introduce the art, science, and ethics of experimentation, debate the validity and utility of experiments as a tool of evaluation and as the basis for policymaking, and examine the findings of experimental research in several distinct political domains. Representative topics include:

1. Voter mobilisation

2. Campaign message testing

3. Social media

4. International development

5. Public health

6. Polling

7. Small-group deliberation

8. Policy nudges

10 hours of lectures and 9 hours of classes in the MT. 1 hour of classes in the ST.

There will be no lecture during Week 6 and no class during Weeks 1 and 6. Reading week activities will include one-on-one meetings with the instructor to discuss final paper topics/proposals (and other course content) and set aside time for students to prepare formative presentations (which will occur in class sessions during Weeks 7-11).

Students will be expected to produce 1 problem sets and 1 presentation in the MT.

The problem set will be due at the end of Week 5 and the presentations will occur during Weeks 7 to 11.

Required textbook:

Glennerster and Takavarasha. 2013. Running Randomized Evaluations: A Practical Guide. Princeton, NJ: Princeton.

Suggested texts:

Desposato. 2016. Ethics and Experiments: Problems and Solutions for Social Scientists and Policy Professionals. New York: Routledge.

Druckman et al., eds. 2012. Cambridge Handbook of Experimental Political Science. New York: Cambridge.

Gerber and Green. 2011. Field Experiments. New York: Norton.

Morgan and Winship. 2015. Counterfactuals and Causal Inference. New York: Cambridge.

Journal articles and other reading list texts, as assigned.

Students will have free choice on final essay topics, so they should expect to find and read additional books and articles on the topic in consultation with the instructor and the library.

Exam (50%, duration: 1 hour and 30 minutes) in the main exam period.
Essay (50%, 2250 words) in the LT.

Assessment for the course will consist of a written exam (1 hour and 30 minutes) during the regular exam period (50%) and an individual case study essay (50%). The essay can take one of two forms: (a) a research design proposal that advances an original theoretical contribution to a political science literature, proposes an experimental test of that theory, and critically discusses related issues of research design, ethics, and implications; or (b) a case study that examines a real-world use of experimental methods to study a question relevant to political science by a government, firm, NGO, or other entity, describes the context for and design of the experiments, and evaluates the findings and use of experiments in this case context.

Students have complete topical choice for the essay and will be expected to give a short presentation (one part of the formative assessment) on their chosen topic during class. Regardless of the form of essay chosen, the quantitative problem set will ensure that students have the skills necessary to propose a credible experimental design and/or evaluate an applied example of experimental research. One-on-one meetings during Reading Week will provide a checkpoint to obtain early feedback and presentations will invite both peer and instructor feedback near the end of the term.