Survey Experiments

Overview

Some of the techniques we discussed last week to reduce sensitivity bias were survey experiments. A survey experiment uses random assignment to administer different pieces of information to respondents, and then compares changes in response to a common outcome. This week, we formalize the potential outcomes framework as a way to establish functional relationships in a model. This framework is useful to understand what we can learn in experiments, survey experiments included.

The lab explores the properties of some commonly used survey experimental designs.

Reading

• RD section 10

• Tomz, Michael R. and Jessica L. P. Weeks. 2013. “Public Opinion and the Democratic Peace.” American Political Science Review 107 (4): 849-865

• Eggers, Andrew C., Nick Vivyan, and Markus Wagner. 2018. “Corruption, Accountability, and Gender: Do Female Politicians Face Higher Standards in Public Life?” Journal of Politics 80 (1): 321-326

Slides

Navigation

Click on the frame below and use the arrows to navigate. You can type ? (or shift + /) to see useful commands.

Open in a new window Download PDF

Lab

Survey experimental designs

Due

Monday, February 12, 11:59 PM