20607 - METHODS AND TOOLS FOR POLICY ANALYSIS

The course will introduce students to the main tools used for data analysis and applied empirical research, focusing on identifying and estimating causal effects. Almost any work in empirical economics (and social science, in general) is about questions of cause and effect such as: Which are the economic returns of one additional year of schooling? Do democratic institutions promote economic development? Does imposing a female policy-maker through gender quotas cause a change in policy? Does raising the minimum wage cause unemployment to increase? Does listening to hate-speech on the radio make people more likely to participate to a genocide? Do longer prison sentences deter crimes? While one would ideally run a controlled experiment to answer these questions, this is often not possible. Therefore, special methods and techniques have been developed in social science research. The mission of the course is to provide background on issues that arise when analyzing social science data and a guide for tools that are useful for applied research. By the end of the course, students should have a firm grasp of the types of research design that can lead to convincing analysis and be able to go through the multiple stages of empirical research: searching for interesting questions, devising an appropriate research design, collecting the data, and implementing the analysis. The course includes practical sessions.

-          The ideal experiment and the potential outcomes framework

-          The simple linear model

-          Instrumental variables

-          Randomized controlled trials and natural experiments

-          Panel data: fixed effects, Difference-in-Differences, synthetic control models

-          Regression Discontinuity Design

-          Additional topics (time permitting)

• Understand the main econometric methods used in empirical research
• Identify the basic properties of estimators and the conditions under which they apply

• Device a research design suitable for a given research question.
• Apply statistical software to conduct regression analyses.
• Interpret and present the findings of econometric analysis.
• Critically engage with texts and journal articles which involve empirical work, recognizing the problems encountered when dealing with data in practice.

• Face-to-face lectures
• Guest speaker's talks (in class or in distance)
• Individual assignments

The learning experience in this course includes traditional lectures and class discussions. The course syllabus will contain information on required readings, including a number of research papers (they should be read in advance so as to facilitate active participation).

Some lectures will be held by invited guest speakers, presenting their empirical research papers [on topics in political economics and political science]. The talks will be followed by a discussion of empirical challenges faced when answering the research question and how they were addressed. This will allow students to better grasp the practical applications of the empirical tools learned in class.

Students will receive individual assignments covering the main topics in the syllabus (4-5 assignments). Students can work on individual assignments with others, but to receive credit every student should hand in a hard copy of an own homework by the stated deadline. The assignment must clearly reflect your own thinking (sets of verbatim copies of homework will not receive any credit).

In order to evaluate the acquisition of the above-mentioned learning outcomes, the assessment of attending students comprises two components:

1.         Written exam (80% of the final grade). The exam can either be taken in two partial exams (counting 40% each) or in one final exam at the end of the course.

2.         Individual home assignment (20% of the final grade). The assignments develop students ability to apply the methods taught during the course in practical situations emerging in data analysis.

In order to evaluate the acquisition of the above-mentioned learning outcomes, the assessment of non-attending students will be based on the written exam (100%).

• Angrist, J., and S. Pischke. Mostly Harmless Econometrics (Princeton University Press, 2008).

Additional textbooks and readings will be indicated in the detailed syllabus.

In some of the assignments students will be asked to solve problems in Stata. A useful reference for applications in Stata is the following:

• Cameron, C. and P.K.Trivedi. Microeconemetrics Using Stata, Revised Edition (Stata Press 2010)

• Angrist, J., and S. Pischke. Mostly Harmless Econometrics (Princeton University Press, 2008).

Additional textbooks and readings will be indicated in the detailed syllabus.