How to Make Causal Inferences with Time-Series Cross-Sectional Data under Selection on Observables

View Researcher's Other Codes

Disclaimer: The provided code links for this paper are external links. Science Nest has no responsibility for the accuracy, legality or content of these links. Also, by downloading this code(s), you agree to comply with the terms of use as set out by the author(s) of the code(s).

Authors Matthew Blackwell, Adam Glynn
Journal/Conference Name AMERICAN POLITICAL SCIENCE REVIEW
Paper Category
Paper Abstract Repeated measurements of the same countries, people, or groups over time form the foundation of many fields of quantitative political science. These measurements, sometimes called time-series cross-sectional (TSCS) data, can help researchers answer a variety of causal questions. Repeated measurements, however, can also lead to confusion about what causal question scholars are answering and what methods, data, and assumptions they need to do so. In this paper, we apply the developments in the statistical literature on causal inference to standard TSCS models and clarify how to nonparametrically define and identify certain TSCS quantities of interest within this context. The paper then describes a number of estimation strategies for these quantities, including inverse probability weighting and structural nested mean models. We show that some of these models will, under strong conditions, be equivalent to some traditional econometric models for TSCS data. This result connects two disparate methodological literatures and shows that some traditional TSCS methods can have a valid interpretation in counterfactual/potential outcomes models. We demonstrate these approaches through two empirical examples. ∗We are grateful to Neal Beck, Jake Bowers, Patrick Brandt, Simo Goshev, and Cyrus Samii for helpful advice and feedback. Any remaining errors are our own. †Department of Government and Institute for Quantitative Social Science, Harvard University, 1737 Cambridge St, ma 02138. web: http://www.mattblackwell.org email: mblackwell@gov.harvard.edu ‡Department of Political Science, Emory University, 327 Tarbutton Hall, 1555 Dickey Drive, Atlanta, ga 30322 email: aglynn@emory.edu
Date of publication 2018
Code Programming Language R
Comment

Copyright Researcher 2022