Introduction: Meta-analysis (MA) is a relatively new method for synthesizing results from a body of empirical studies. MA can answer certain theoretical and practical questions that primary studies cannot due to pragmatic factors. Recent SSWR conferences have included activities designed to increase understanding of MA and promote its use. However, little is known about the degree to which Social Work and allied disciplines use MA. This study compares production of MA-related studies across five disciplines. The results can guide Social Work's future steps with regard to MA.

Method: Five disciplines that promote the human condition were compared: Family Studies, Nursing, Psychology, Psychiatry, and Social Work. We limited our search to MA articles in peer-reviewed journals. Journals were chosen based on their impact rating (Journal Citation Reports of Social Sciences Edition Journal Rankings, 2002). We selected the 10 journals with the highest impact ratings from each discipline. Thus, 50 journals were reviewed.

All abstracts containing the term “meta” were obtained and coded into one of four categories: actual MA, methodology of MA, review of a previous MA, or false positive. Abstracts were organized into five time frames: 1980–1984, 1985–1989, 1990–1994, 1995–1999, and 2000–2004. Additionally, the total number of articles per journal within each time frame was obtained to calculate the percentage of journal contents involving MA articles. Abstracts were independently coded by two research assistants (Kappa = .86).

Results: As predicted, the number of published MA increased over time across each discipline, r = .98. Family Studies, Nursing, and Psychiatry had correlations exceeding r = .95, whereas Social Work and Psychology had lower correlations, rs = .80 and .79. The relative percentage of MA articles also increased across time for each discipline, r = .93. When considered separately, all but Family Studies (r = .07) and Social Work (r = .51) had correlations above .89. Graphical representations indicate that methodological articles on MA increased between 1980 and 1995 and have since decreased, suggesting the MA methodology has matured. ANOVAs were conducted to determine if disciplines differed in publication of MA scholarship. Significant differences were found for both the number of and relative percentage of actual MA and methodology articles, all Fs (4, 20) > 5.70, ps < .005. Post-hoc analyses indicate that Psychology has led production of MA-related scholarship. However, Psychiatry has published more MA than Psychology since 1995. Social Work has consistently produced the least MA-related scholarship. Given that the two disciplines publishing the most MA also enjoy significantly higher publication impact ratings, Social Work might consider producing more MA.

Discussion: MA reviews can strengthen theory and inform Social Work practice decisions. Our results suggest that Social Work should continue to promote MA. Such efforts should focus less on methodological issues related to MA and more on producing actual MA, and on ensuring that today's primary studies contain sufficient data for tomorrow's quantitative reviews.