The effects of social support: A meta-analysis of studies that used the Norbeck Social Support Questionnaire




Neatherlin, Jacquelin

Journal Title

Journal ISSN

Volume Title



There has been a myriad of literature and research on the concept of social support. This meta-analysis study examined nursing research studies that used the Norbeck Social Support Questionnaire (NSSQ) to measure the effect of social support on outcome variables. This questionnaire is based on Kahn's conceptualization of social support. There were three research questions to guide the study relating to: (a) the magnitude of the effect of social support on outcome variables, (b) the difference among the components of support in regard to the magnitude of their effects on outcome variables, and (c) the demographic variables associated with the individual studies.

Fifty-two studies were reviewed for the meta-analysis, with 13 being initially eliminated. Three studies were later excluded from further data analysis for statistical reasons. One study was coded three times due to the significant changes in sample size over the three different data collection time periods. The studies were published or completed between 1983 and 1991.

Thirty-eight studies from 36 sources, representing 3,272 subjects, were included in the meta-analysis. There were 217 individual effect sizes and 15 outcome variable categories. The unweighted r mean was.178 for the 217 individual effect sizes, and the mean weighted r was.155 for the 15 outcome variable categories, which included 7 functional, 4 network, 1 loss, and 3 total categories.

All of the effect sizes for the outcome variable categories were either small or essentially zero. The largest effect size was the relationship between network and life perceptions (r =.282). The smallest effect size was the relationship between total support and job factors (r = .06). There was no significant difference among the NSSQ components of social support in the magnitude of their effects on selected outcome variable categories. The demographic variables indicated that most of the sample sizes contained 100 subjects or less, subjects generally were female, Caucasian, and high school or college-level educated. Conclusions reached were that social support effects were generally small and that there may be confusion as to the proper use of the NSSQ.



Nursing Society Questionnaires, Meta-analysis, Norbeck Social Support Questionnaire