NCJ Number
176726
Date Published
1997
Length
24 pages
Annotation
This paper introduces research synthesis methods for the analysis of complex processes and outlines how they can be applied in the study of the literature on substance abuse prevention.
Abstract
The idea of synthesizing available information about treatment efficacy or the strength of relationships among variables is not new; however, meta-analysis has been criticized for attending only to main effects (Cook and Leviton 1980) and ignoring the important roles of mediating and moderating variables. Recently, reviewers in a number of disciplines have realized that research synthesis techniques can be applied in reviews of issues more complex than those previously studied. In a discussion of model-driven meta-analysis, this paper describes a model for the roles of risk and protective factors in substance abuse prevention, based on the review of Hawkins and colleagues (1992). This model contains 11 broad predictors of substance use and abuse outcomes. Contextual factors are laws, drug availability, norms, socioeconomic status, and neighborhood disorder. Interpersonal and individual factors encompass parental values and behaviors, individual values and behaviors, and peer values and behaviors. The model shows 19 paths or connections among components. Both direct and indirect influences are indicated. Following a description of the model, the author next discusses how evidence about models could be obtained and examined in a quantitative synthesis of the literature on this topic; key issues that arise in the application of this approach are also described. A brief example of data analysis for a four- variable model is also presented. The paper concludes with a discussion of how a model-based synthesis of risk and protective factors could be used in the design and analysis of substance abuse prevention programs. 4 tables, 3 figures, 2 notes, and 28 references