Causal Inference in Sociological Research

Abstract

Originating in econometrics and statistics, the counterfactual model provides a natural framework for clarifying the requirements for valid causal inference in the social sciences. This article presents the basic potential outcomes model and discusses the main approaches to identification in social science research. It then addresses approaches to the statistical estimation of treatment effects either under unconfoundedness or in the presence of unmeasured heterogeneity. As an update to Winship & Morgan's (1999) earlier review, the article summarizes the more recent literature that is characterized by a broader range of estimands of interest, a renewed interest in exploiting experimental and quasi-experimental designs, and important progress in the areas of semi-and nonparametric estimation of treatment effects, differencein-differences estimation, and instrumental variable estimation. The review concludes by highlighting implications of the recent econometric and statistical literature for sociological research practice.

Can a Flexibility/Support Initiative Reduce Turnover Intentions and Exits? Results from the Work, Family, and Health Network

“…Increasingly sociologists and other scholars are proposing randomized experiments as the gold standard for understanding social processes (Gangl 2010;Morgan and Winship 2007;Winship and Morgan 1999). Others underscore the importance of locating lives in context to better specify the conditions under which some processes operate (Brooks-Gunn et al 1993;Clampet-Lundquist et al 2011;Ludwig et al 2008;Morenoff 2003;Sampson 2008).…”

Using Factorial Surveys to Study Justice Perceptions: Five Methodological Problems of Attitudinal Justice Research

“…Empirical justice research that intends to provide not only mere descriptions of collective opinions and attitudes but also explanations on the basis of theoretically derived statements on causal connections is confronted with the same problem that all empirical social research has, namely that of modeling causal relationships and of sufficiently testing these relationships using empirical methods (Gangl, 2010;Opp, 2010). Traditional survey-based research, which uses cross-sectional data, is practically incapable of reliably identifying possible causal relationships, since correlations between two variables might just as well be caused by other variables that have not been measured.…”

“…Traditional survey-based research, which uses cross-sectional data, is practically incapable of reliably identifying possible causal relationships, since correlations between two variables might just as well be caused by other variables that have not been measured. Due to the simultaneous measurement of theoretically assumed causes and effects, and due to the problem of unobserved heterogeneity, complex methods are needed to identify causal relationships post hoc (Gangl, 2010). Although longitudinal studies can be used to measure causes and effects separately in time, and appropriate methods (fixed-effects models) can be used to exclude the timeconstant unobserved heterogeneity (Allison, 2009;Brüderl, 2010), 10 the best way to test causal relationships is to use experimental methods (Falk & Heckman, 2009).…”

Social Movements, Risk Perceptions, and Economic Outcomes

“…Endogeneity is typically a concern when the dependent variable has a potential causal effect on the independent variables in question. This necessitates isolating the causal effect of the independent variable through the use of an instrument (Gangl 2010). We created instrumental variables with two-stage least square regression, using the xtivreg command in Stata 12, with random effects.…”