Meta-Analysis and Social Services Interventions

Meta-Analysis and Social Services Interventions

DOI: 10.4018/978-1-7998-1147-3.ch002
OnDemand PDF Download:
No Current Special Offers


The percentage of recidivism reduction projected to reduce recidivism by each social service or intervention is presented. Meta-analysis is used to determine these projections. However, in the last few years meta-analysis methods have been questioned. The most formidable criticism of meta-analysis is found in a master study of the statistical power of studies in criminology. This master study (of over 8,000 studies) found that about 25% of studies in criminology exhibited high statistical power (in the 0.99 to 1.00 range). However, the study also found about 25% of studies had power between 0.01 and 0.24. These findings suggested that roughly a fourth of all studies in criminology have levels of statistical power that make it nearly impossible to identify the effects they are estimating. In other words, this chapter questions whether we should be confident in the recidivism reduction projections for various interventions.
Chapter Preview

Meta-analysis is widely used in the social, behavioral, and medical sciences to combine results of multiple studies and produce relevant information for clinical practice and social policy. It is most often used to synthesize quantitative data on treatment effects, but it has many potential applications. Meta-analysis includes a set of techniques for quantitative data synthesis that can (and should) be performed in the context of systematic efforts to minimize bias at each step in the research review process (see Systematic Reviews). Without careful efforts to eliminate bias, meta-analysis can lead to wrong conclusions. (Encyclopedia of Social Work, n.d.).


The Steps To Performing Meta-Analysis

The basic steps to meta-analysis are the following:

  • 1.

    Select the independent (causal) and dependent (outcome) variables of interest.

  • 2.

    Locate all relevant and usable studies containing information about the effect of interest. Regarding this step:

    • Guzzo, Jackson, and Katzell (1987) describe numerous judgment calls that must be made by the meta-analyst. These judgment calls introduce considerable subjectivity. Are technical reports in or out? Which disciplines are likely to produce relevant dissertations? Does the age of the study matter? Does the language in which the report is written matter? Which journals will be searched?

    • Although computer searches are more convenient than manual searches, they are not more thorough or accurate than manual searches. A search is completely dependent on the keywords used to perform the search. Thus, it is virtually impossible to estimate how many studies containing relevant data are likely to be passed over by an electronic search because they are not keyworded.

  • 3.

    Code the characteristics of each study that might relate to the size of effects obtained in different studies.

  • 4.

    Calculate effect size estimates for independent-dependent variable pairs of interest. An effect size, called a d-value, is obtained by subtracting the control (or comparison) group's mean on the dependent variable from the experimental group's mean on the dependent variable and dividing by the standard deviation for the control group. Specifically, d=(Xe-X`)SDc.

  • 5.

    Calculate the mean effect size(s) across studies. The calculation of effect-size estimates requires information about sample size, a commonly reported datum, and information about variances, which are often not reported (Guzzo, Jackson & Katzell, 1987).

(Excerpts from Glass, McGaw, and Smith, 1981)

Andrews and Bonta (2010) are widely cited for their expertise in using meta-analysis to study social recidivism reduction service interventions and their preference for the method is indicated in the following quote, “The traditional literature has been narrative in nature and the qualities of the review depend very much upon the expertise and thoroughness of the author(s). … Thus, it is not uncommon for two independent reviews of a particular literature to reach very different conclusions.” They go on to argue “Meta-analytic reviews permit a more unbiased analysis of the literature and they provide aquantitativeanalysis estimate of the importance of the results” (Andrews & Bonta, 2010, p. 32,).

The following section presents a concise summary of much of the results of meta-analytic research within the realm of recidivism reduction interventions. The intervention’s name is on the left-hand side of the table, the authors that did the research are in the middle of the table and the percent of recidivism reduction that can be expected is on the right-hand side of the table.


Programs / Services And Recidivism Reduction

Table 1 indicates the major studies for each social service intervention and the percent of recidivism reduction (or increase) expected based on the meta-analysis research strategies.

Complete Chapter List

Search this Book: