Suppressor Variables in Social Work Research: Ways to Identify in Multiple Regression Models

When selecting a set of study variables, social work researchers frequently test correlations between the outcome variables (i.e., dependent variables) and theoretically relevant predictor variables (i.e., independent variables). In some instances, one or more of the predictor variables are uncorrelated with the outcome variable. This situation poses the question of whether researchers’ multiple regression analyses should exclude independent variables that are not significantly correlated with the dependent variable. Questions such as this are routine, and our article provides a systematic answer to these questions. In the multiple regression equations, suppressor variables increase the magnitude of regression coefficients associated with other independent variables or set of variables (Conger, 1974). A suppressor variable correlates with other independent variables, and accounts for or suppresses some outcome-irrelevant variation or errors in one or more other predictors, and improves the overall predictive power of the model. Given this function, some prefer to call the suppressor variable an enhancer (McFatter, 1979). A variable may act as a suppressor or enhancer—even when the suppressor has a significant zero-order correlation with an outcome variable—by improving the relationship of other independent variables with an outcome variable. This type of suppressor variable is more likely to be retained in a regression model than a variable that has a zero correlation with the outcome variable. However, this article aims to underscore the value of retaining