Summary Many studies have reported positive correlations between dream recall frequency (DRF) and measures of absorption, psychological boundaries and attitude towards dreams. A majority of these studies, however, have relied exclusively on retrospective measures of DRF even though daily dream logs are generally considered to be more direct and valid measures of DRF. The first goal of the present meta-analysis was to evaluate the effect sizes of three variables (absorption, psychological boundaries and attitude towards dreams) as correlates of DRF. The second goal was to evaluate if these effect sizes varied as a function of how DRF was operationalized (i.e. retrospective measure versus dream log). Data from 24 studies were included in the analyses. For each of the three variables investigated, correlations with retrospective measures of DRF were of greater magnitude than those obtained with daily logs. These results indicate that scores on measures of absorption and psychological boundaries are not related to DRF per se, but rather to people's tendency to retrospectively underestimate or overestimate their DRF, while attitude towards dreams is related both to DRF per se and to people's retrospective estimation bias. Implications of these findings for dream research are discussed.