Читать книгу The Research Experience - Ann Sloan Devlin - Страница 72

A meta-analysis is sometimes referred to as a study of other people’s studies or “an analysis of analyses” (Glass, 1976, p. 3). Another definition is the “Integration of the findings of several research studies by means of statistical techniques focusing on the same research question leading to meaningful quantitative data” (Corsini, 2002, p. 588). Results of meta-analyses combine the data from studies that are asking the same research question to produce a result that tells you how strong the impact of the variable of interest is. The result is reported as an effect size. An effect size is a quantitative indicator of the strength of a particular occurrence. For example, a company might be interested in having onsite parking at work and wants to know whether having an onsite parking garage is associated with lower absenteeism than having an offsite parking garage that requires using a shuttle bus to get to the office. If you were to conduct a meta-analysis on this issue, you would use appropriate search terms, in a comprehensive list of databases, to find research where onsite versus offsite parking was examined and one of the outcomes was a measurement of absenteeism. There are specific steps you would then take to combine the results of the studies you found numerically to determine the effect size.

Meta-analysis: Meta-analysis is a “study of studies” that uses a statistical approach to synthesize the findings on a particular topic and to report the impact of a given intervention. That impact is reported as an effect size.

Effect size: Quantitative indication of the strength of a particular occurrence.

One value of a meta-analysis is that it includes a systematic review of the relevant research by the authors. The research studies that were included to calculate the effect size are included in the published meta-analysis, giving you a solid list of references. When you look at a meta-analysis, you get a pretty good idea of the research conducted on this topic. But it is wise to remember another saying sometimes associated with meta-analyses: “garbage in–garbage out,” a phrase that appears to have originated in the computing world (http://www.worldwidewords.org/qa/qa-gar1.htm). What that phrase means is that the scientific rigor of the studies included in the meta-analysis is important. If the studies are not of high quality, then the outcome of the meta-analysis is questionable.