Skip to Main Content
Have library access? Log in through your library
Meta-Analysis for Explanation

Meta-Analysis for Explanation: A Casebook

Thomas D. Cook
Harris Cooper
David S. Cordray
Heidi Hartmann
Larry V. Hedges
Richard J. Light
Thomas A. Louis
Frederick Mosteller
Copyright Date: 1992
Published by: Russell Sage Foundation
Pages: 392
  • Cite this Item
  • Book Info
    Meta-Analysis for Explanation
    Book Description:

    Social science research often yields conflicting results: Does juvenile delinquent rehabilitation work? Is teenage pregnancy prevention effective? In an effort to improve the value of research for shaping social policy, social scientists are increasingly employing a powerful technique called meta-analysis. By systematically pulling together findings of a particular research problem, meta-analysis allows researchers to synthesize the results of multiple studies and detect statistically significant patterns among them.

    Meta-Analysis for Explanationbrings exemplary illustrations of research synthesis together with expert discussion of the use of meta-analytic techniques. The emphasis throughout is on the explanatory applications of meta-analysis, a quality that makes this casebook distinct from other treatments of this methodology. The book features four detailed case studies by Betsy Jane Becker, Elizabeth C. Devine, Mark W. Lipsey, and William R. Shadish, Jr. These are offered as meta-analyses that seek both to answer the descriptive questions to which research synthesis is traditionally directed in the health and social sciences, and also to explore how a more systematic method of explanation might enhance the policy yield of research reviews.

    To accompany these cases, a group of the field's leading scholars has written several more general chapters that discuss the history of research synthesis, the use of meta-analysis and its value for scientific explanation, and the practical issues and challenges facing researchers who want to try this new technique. As a practical resource,Meta-Analysis for Explanationguides social scientists to greater levels of sophistication in their efforts to synthesize the results of social research.

    "This is an important book...[it is] another step in the continuing exploration of the wider implications and powers of meta-analytic methods." -Contemporary Psychology

    eISBN: 978-1-61044-133-9
    Subjects: Sociology, Psychology

Table of Contents

Export Selected Citations Export to NoodleTools Export to RefWorks Export to EasyBib Export a RIS file (For EndNote, ProCite, Reference Manager, Zotero, Mendeley...) Export a Text file (For BibTex)
  1. Front Matter
    (pp. i-iv)
  2. Table of Contents
    (pp. v-vi)
  3. Foreword
    (pp. vii-x)
    Eric Wanner

    The practical value of social science depends upon its ability to deliver usable knowledge about the causes of social problems and the effectiveness of policies and programs designed to alleviate them. The immense diversity of social life, however, and the great welter of factors underlying any social phenomenon make it difficult, if not impossible, to derive conclusive knowledge from any single study, no matter how well designed or intelligently analyzed. The causal process that appears so essential at one time or place may prove less important in another. The program that works well with one group under certain conditions may...

  4. Annotated Bibliography of Meta-Analytic Books and Journal Issues
    (pp. xi-xiv)
    Richard Light and Frederick Mosteller
  5. 1 The Meta-Analytic Perspective
    (pp. 1-16)

    Social science seeks to produce descriptions and explanations of phenomena that apply across a wide range of persons and settings. The hope is that such knowledge will improve both substantive theory and public policy. Policymakers particularly value descriptive knowledge if it can be generalized to many population groups and settings because they need to identify successful practices that can be transferred to a wide range of settings. Such knowledge transfer is enhanced if policymakers also know why relationships come about. Knowing which processes bring a desired end makes it easier to design new programs or modify existing ones in ways...

  6. 2 Explanation in Meta-Analysis
    (pp. 17-34)

    Explanation entails understanding how or why something happens as opposed to describing what has happened. We can describe changes in academic achievement levels, but to explain why achievement rises or falls over time we must somehow take into account those antecedents that produce or generate the observed variation. Or we might, by means of randomized experiments and valid measures, determine whether a particular type of patient education is associated with longer or shorter hospital stays for particular population groups. But to do so would not necessarily imply anything about how or why one type of education is effective for some...

  7. 3 Effects of Psychoeducational Care with Adult Surgical Patients: A Theory-Probing Meta-Analysis of Intervention Studies
    (pp. 35-82)
    Elizabeth C. Devine

    There is a large research base of controlled, clinical trials on the effects of patient education and/or psychosocial support (hereafter called psychoeducational care) administered to adult surgical patients. Initial metaanalyses of this research (Devine and Cook 1983, 1986) have demonstrated three important findings.

    First, patients receiving additional psychoeducational care recovered more quickly, experienced less postsurgical pain, had less psychological distress, and were more satisfied with the care they received than patients receiving the psychoeducational care usually provided in the setting. For these four outcomes, averagedvalues, based on a sample of studies, were .50 (n= 73), .39 (n...

  8. 4 Juvenile Delinquency Treatment: A Meta-Analytic Inquiry into the Variability of Effects
    (pp. 83-128)
    Mark W. Lipsey

    One need not look beyond the daily newspaper to establish that crime is a matter of considerable concern in our society. Far less obvious is what should be done about it. As with almost any important matter, this is one on which opinions can differ sharply, not only in the political arena but among social science researchers and criminological experts as well.

    Among the many approaches to crime prevention that have been advocated are punishment (deterrence), amelioration of social conditions that produce crime, target hardening, community prevention (e.g., neighborhood watches), and a host of other such notions. Of particular interest...

  9. 5 Do Family and Marital Psychotherapies Change What People Do? A Meta-Analysis of Behavioral Outcomes
    (pp. 129-208)
    William R. Shadish Jr.

    What best distinguishes family and marital therapies from other therapies is the belief that the crucial forces in a person’s life are his or her interactions with family members (Nichols 1984). Thus family members are usually included in therapy to help change family interaction patterns.

    Perhaps the most common way to categorize different kinds of marital and family therapies is through their theoretical orientation; that is, the system that therapists use for guidance in dealing with clients. Much controversy exists in the psychotherapy literature about whether orientation makes any difference to therapy outcome. Examining the magnitude of orientation effects, and...

  10. 6 Models of Science Achievement: Forces Affecting Male and Female Performance in School Science
    (pp. 209-282)
    Betsy Jane Becker, Christine M. Schram, Lin Chang, Mary M. Kino and Maria Quintieri

    For the past decade scientists and science educators around the world have regarded the small numbers of women in scientific careers as a critical problem (e.g., Bruer 1983; Science Council of Canada 1981; National Research Council 1983). Although women’s levels of participation in scientific enterprises have increased over the past two decades (see, e.g., National Research Council 1983, p. 16), the increases have not occurred uniformly throughout all of science. In the United States, for instance, the numbers of women pursuing education in the physical sciences have been very small, whereas in biology and other life sciences almost 30 percent...

  11. 7 Some Generic Issues and Problems for Meta-Analysis
    (pp. 283-320)

    The studies included in this volume reveal that meta-analysis, like any other research endeavor, has its own internal logic. As described in Chapter 1, the organizational structure can be seen in Cooper’s model (1989) of the stages of meta-analysis. The case illustrations also reveal a level of flexibility in the meta-analytic process. For example, literature searches (part of the data collection phase) may reveal that we have too few studies to answer the initial questions specified as part of the problem formulation. Problem formulation is likely to be iterative and can become more clearly focused over the course of the...

  12. 8 What Have We Learned About Explanatory Meta-Analysis?
    (pp. 321-342)

    In this final chapter we examine how the eight explanatory tasks set out in Chapter 2 were handled in the cases described in this volume. To reiterate, these tasks require identifying (1) those mediating processes that causally link one construct to another, particularly a cause and an effect; (2) those components of a treatment responsible for influencing a particular outcome; (3) those components of an outcome that have been impacted by a causal agent; (4) those person, setting, and time variables that moderate a descriptive causal relationship; (5) the treatment classes that influence an outcome; (6) the theoretical integrity of...

  13. References
    (pp. 343-366)
  14. Index
    (pp. 367-378)