A meta-analysis of the effects of experimental manipulations of social support during laboratory stress

A meta-analysis was conducted on 22 experimental studies of the effects of social support on reactivity to laboratory stress. The analysis included 26 comparisons and 56 effect sizes, with the number of participants in each comparison ranging from 20 to 104. The average effect size was 0.61 for heart rate and systolic blood pressure, 0.51 for diastolic blood pressure, 0.25 for skin conductance, and 0.83 for cortisol. It was concluded that the laboratory analogues examined to date provide tentative confirmation of the hypothesis that social support has a protective health effect. Recommendations are made for future research, including the need to improve statistical power; expand subject participation beyond students; and increase consistency between studies in relation to measurements of social support effects, methods for manipulating social support, and types of behavioral challenges and control procedures used.