Overview of Methods for Adjustment and Applications in the Social and Behavioral Sciences: The Role of Study Design

In this chapter, the authors provide an overview of methods to adjust for factors that may confound the relationship between the exposure and outcomes of interest, including discussion of some of the particular challenges that emerge in the social and behavioral sciences. Two common covariate summary measures are the Mahalanobis distance and the propensity score. Weighting adjustments typically involve direct use of the estimated propensity scores to assign weights for each individual. The most common analysis-based method involves constructing an outcome model with the exposure indicator and the observed covariates as predictors. In non-experimental studies, adjustment methods are used to induce comparability between the exposed and unexposed. In short, different choices and assumptions may need to be made depending on the scientific context and the data that is typically available in that context, and it is crucial to assess the underlying assumptions within the context of each specific study.