SOS9019 – Research Designs for Causal Inference
Course content
This course offers an introduction to several research designs particularly suited to making causal inferences. Social scientists routinely ask causal questions such as “Does divorce affect children’s school performance?” and “Does family policies affect gender relations?” Questions of this type are notoriously difficult to answer, and many sociologists have traditionally been very careful not to voice any causal ambition on behalf of their empirical analyses.
Researchers in the social sciences now have at their disposal a wide array of tools that allows them to provide more plausible answers to causal questions. In this course, participants will be introduced to and trained in a selection of the most commonly used and most powerful of these tools.
Taking the counterfactual model of causal effects as the point of departure, the problems associated with drawing causal inferences from observational data will be discussed. From these, the course moves on to demonstrations of various research designs that are particularly well suited for establishing causal effects. These include regression discontinuity designs, panel data methods including «differences-in-differences», and instrumental variable models.
The course is applied, in the sense that it is taught through a combination of lectures and computer lab sessions where participants will replicate results from exemplary studies. Time will also be allotted to critically discuss empirical challenges from the participants’ own research projects.
Admission
This course is aimed at doctoral students, researchers and talented master students in the social sciences with an interest in causal inference. 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.
Participants outside the Department of Sociology and Human Geography shall fill out this application form.
The application deadline is four weeks prior the course.
Spring 2017 the application deadline is 13th February 2017!
Overlapping courses
SOS4021 Research Designs for Causal Inference (Master-level).
Teaching
This course has jointly taught classes with SOS4021 Research Designs for Causal Inference (master-level). The course will be organized as lectures, seminars, and labs.
For detailed schedule please go to the semester page "Spring 2017" which you find above.
Instructors
Andreas Kotsadam holds a Ph.D. from the University of Gothenburg. Niklas Jakobsson holds a Ph.D. from University of Gothenburg. Both publish in several social science disciplines.
Examination
Participants on Ph.D.-level obtain 6 ECTS credits by completing the course requirements, which are active participation in the course, submission of a homework (referee reports and an applied problem set) and presentation of a research proposal.
Master students find detailed information about course requirements on the webpage of the master-variant of this course - SOS4021. Please go to this page for further information as requirements for master-students are slightly different from those for Ph.D.-students..
Grading scale
Grades are awarded on a pass/fail scale. Read more about the grading system.