We evaluated publication bias in several ways. First, we utilized the “fail-safe N” method (Rosenthal, 1991; Rosenthal and Rubin, 1988) to determine the number of additional studies with a null result needed to reduce the overall effect size to non-significance. Rosenthal (1991) suggested that if the fail-safe N exceeds 5 multiplied by K (the number of studies in the meta-analysis) + 10, the results of the meta-analysis can be interpreted as robust. Second, we conducted a visual inspection of the funnel plot to evaluate symmetry relative to the mean effect size (with greater symmetry reflective of a lesser likelihood of publication bias). Finally, year of publication was evaluated as a potential moderator. Analyses were completed using the standard software program Comprehensive Meta-Analysis (Version 2; Borenstein et al., 2005).
