Integrating Findings: The Meta-Analysis of Research

Abstract

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. American Educational Research Association is collaborating with JSTOR to digitize, preserve and extend access to Review of Research in Education.Maccoby and Jacklin's (1974) review of research on the psychology of sex differences encompassed 1,600 works published before 1973. Considering the literature on that topic since 1973, and the fact that many other studies that are not focused specifically on sex differences may contain data on the question, a population of over 5,000 studies on sex differences can be imagined. There are dozens of educational problems on which the research literature is comprised of several hundred articles: ability grouping, reading instruction, programmed learning, instructional television, integration, and so on.Educational research and evaluation is a large and widely scattered enterprise. On problems of importance, it produces literally hundreds of studies in less than five years. The research techniques used, the measurements taken, the types of person studied-all may vary from one study to the next, even though the topic is the same. The research enterprise in education and the social sciences is a rough-hewn, variegated undertaking of huge proportions. Determining what knowledge this enterprise has produced on some question is itself a genuinely important scholarly endeavor.The evolution of style of research integration has been shaped by the size of the research literature. In the 1940s and 1950s, a reviewer contributing to the Review of Educational Research or Psychological Bulletin might find one or two dozen studies on a topic. A narrative, rhetorical integration of so few studies was probably satisfactory. By the late 1960s, the research All use subject to JSTOR Terms and Conditions 352 Review of Research In Education, 5 literature had swollen to gigantic proportions. Although scholars continued to integrate studies narratively, it was becoming clear that chronologically arranged verbal descriptions of research failed to portray the accumulated knowledge. Reviewers began to make crude classifications and measurements of the conditions and results of studies. Typically, studies were classified in contingency tables by type and by whether outcomes reached statistical significance.Integrating the research literature of the 1970s demands more sophisticated techniques of measurement and statistical analysis. The accumulated findings of dozens or even hundreds of studies should be regarded as complex data points, no more comprehensible without the full use of statistical analysis than hundreds of data points in a single study could be so casually understood. Contemporary research reviewing ought to be undertaken in a style more technical and statis...

Olfactory Dysfunction in Schizophrenia A Qualitative and Quantitative Review

“…The dependent measure was effect size for tests of olfactory: (i) identification; (ii) detection threshold; (iii) discrimination; and (iv) recognition memory expressed in Cohen's d (Cohen 1977;Glass 1977). The d score is the difference between patient and control group means, within each study or comparison, expressed in standard deviation units.…”

“…The d score is the difference between patient and control group means, within each study or comparison, expressed in standard deviation units. Where the means and standard deviations were not reported, t , F , r , or 2 statistics were converted to d using formulas provided by Glass (1977). By expressing effect size in standard deviation units, we were able to make a direct comparison of outcomes across studies.…”

The effect of psychological intervention on recovery from surgery and heart attacks: an analysis of the literature.

“…A positive ES in the Appendix 'tables signifies the difference favors the group receiving the psychological intervention. 22 …”

The Synthesis of Partial Effect Sizes

“…Meta-analysis is the statistical analysis of empirical results of a series of studies (Glass, 1977). Quantitative procedures in meta-analysis typically begin by transforming study results to a common metric, which often represents results as effect sizes.…”