Fig 4. Overestimation (Y) of the random-effects model when compared with the 10% most precise observed effect sizes for simulated data based on characteristics of subsets from meta-analyses published in Psychological Bulletin (left panel) and Cochrane Database of Systematic Reviews (CDSR; right panel).
pub refers to the extent of publication bias and open bullets, triangles, and diamonds indicate no, small, and medium average true effect size. The solid line indicates the mean of the Y-variable observed in the homogeneous subsets and the dashed lines are the upper and lower bound of a 95% confidence interval (CI) around the mean of the Y-variable.