A meta-analysis of 17 studies was conducted to determine the magnitude of the relationship between postpartum depression and infant temperament during the infant's first year. Interrater reliabilities for coding of substantive and methodological characteristics for each study ranged from 87% to 100%. The meta-analytic combinations were calculated in three different ways: unweighted, weighted by sample size, and weighted by a quality index score. A significant, moderate correlation between postpartum depression and infant temperament was found. The mean r index of effect size was .31 when the studies were weighted by sample size; .36 when the studies were unweighted; and .35 when they were weighted by the quality score. All three r indexes were in Cohen's range of moderate effect sizes. The calculated 95% confidence interval ranged from .261 to .369, indicating a significant relationship between postpartum depression and infant temperament. Sample size and year of publication were significantly correlated with the effect size of temperament in infants of postpartum depressed mothers.
Cheryl Tatano Beck, DNSc, CNM, FAAN, is a professor, College of Nursing, University of Rhode Island, Kingston, RI.
Accepted for publication January 19, 1996.