Abstract
Background: The translation and grounded theory techniques typically cited as the method for producing qualitative metasyntheses do not lend themselves well to qualitative survey findings as they do not contain the integrating concepts or controlling metaphors upon which these techniques depend.
Objectives: The purpose of this article is to describe a process for creating metasummaries of qualitative survey findings.
Methods: This article is based on completed work in an ongoing methodological study aimed at developing a usable and transparent protocol for combining the findings in reports of health-related qualitative studies. The sample for this work included 45 published and unpublished reports of qualitative studies of HIV-positive women with findings on motherhood, 39 of which contained findings in the form of surveys of data. Almost 800 findings were extracted. These extracted findings were reduced to 93 abstracted findings, and manifest frequency and intensity effect sizes were calculated.
Results: Five findings had effect sizes ranging from 25-60%, with both published and unpublished reports contributing about equally to the strength of these findings. Seventy-three findings had effect sizes of <9%, 47 of them with effect sizes of only 2%. In most of these cases, only one work contained the finding and these works had generally fewer space restrictions. Four reports with few space restrictions contained 63% of the findings across all 45 reports.
Conclusions: Qualitative metasummaries are useful end products of research integration studies involving reports of findings in the form of qualitative surveys, and may serve as a foundation for qualitative metasyntheses.
Interest has surged in combining the findings from qualitative studies to enhance their use in evidence-based practices (Green & Britten, 1998;Popay & Williams, 1998;Sandelowski, 1997;Sandelowski, Docherty, & Emden, 1997). This new turn to qualitative metasynthesis is evident in the burgeoning methodological literature on the subject (e.g., Jensen & Allen, 1996;Noblit & Hare, 1988;Paterson, Thorne, Canam, & Jillings, 2001;Sandelowski, Docherty, & Emden, 1997), and in the growing number of studies designated as qualitative metasyntheses or the like (e.g., Barroso & Powell-Cope, 2000;Beck, 2002;Kearney, 2001).
In these publications, grounded theory and Noblit & Hare's (1988) translation techniques are the methods most frequently cited for combining qualitative findings. These techniques depend on findings presented in the form of integrating concepts, models, or controlling metaphors that represent researchers' interpretive syntheses of the data they collected. Yet reports of qualitative studies do not always contain such interpretive syntheses, but instead contain summaries, or surveys, of data organized by topics or themes. The qualitative summaries in these reports may therefore lend themselves better to metasummary than to metasynthesis (Table 1).
The purpose of this article is to describe techniques that can be used to create metasummaries of qualitative findings. This purpose is in the service of: (a) clarifying the difference between qualitative findings in individual research reports in the form of summaries as opposed to syntheses of data; and, of (b) establishing a difference between metasummaries and metasyntheses of qualitative findings as end-products of research integration studies or systematic reviews of qualitative research in a target domain. Arguably, qualitative metasummary is a form of systematic review more amenable to research reports where findings are summaries or surveys of the data researchers collected in the course of their studies. In contrast, qualitative metasynthesis is a form of systematic review more amenable to research reports where findings are syntheses of the data collected. Consistent with the purpose of this article, the results of the metasummary produced are referred to here only to make a point about method (a complete list of references and displays of metasummary results may be found on the Nursing Research Editor's Web site at http://sonweb.unc.edu/nursing-research-editor)