A foundational assumption in meta-analysis is that effect sizes (EFSs) based on scores from different measures are directly comparable. Recent theoretical work has shown that in order for EFSs to be comparable in the sense of being interchangeable across measures, the EFSs must be based on scores meeting two conditions: construct equivalence, and the reliabilities of the scores from these measures must be equal. No research has investigated the consequences for the results of meta-analytic analyses and comparisons if the EFSs are based on scores failing to meet these two conditions.