Methodology

A More Conscious Experimental Design: How More Explicit Use of Process Tracing Can Improve Experiments

Derek Beach Aarhus University
,
Levi Littvay Central European University

Abstract

Randomized experimental design is seen by many researchers in the social sciences as the gold standard of causal inference, but as a method, it is blind to the mechanistic processes that lead from cause to outcome. While experimental researchers are highly conscious of the mechanisms underlying their theorization and inference, they are mostly implicit in their research design and analysis. We argue that process tracing used as an adjunct method alongside an experiment can be used to make the mechanistic assumptions explicit in experimental design and analysis. Through this combination, experimental researchers can become more conscious about causal mechanisms in their theorization, leading to better experimental designs, more effective ways to discuss problems with causal heterogeneity, and even strengthening the internal and external validity of findings.

Content

Thumbnail image of Experiments draft APSA.pdf
cloud_download

Comments

Log in or register with APSA to comment open_in_new
Comments are not moderated before they are posted, but they can be removed by the site moderators if they are found to be in contravention of our Commenting Policy open_in_new – please read this policy before you post. Comments should be used for scholarly discussion of the content in question. You can find more information about how to use the commenting feature here open_in_new .
This site is protected by reCAPTCHA and the Google Privacy Policy open_in_new and Terms of Service open_in_new apply.