Secondary Data Analysis

Similarities and Distinctions Between Secondary Data Analysis and Meta-Analysis

Graves considered meta-analysis to be a special case of secondary analysis. However, because of differences in the two methods, some researchers do not agree with Graves. A comparison of these two methods reveals both similarities and distinctions. Regarding similarities, both methodologies use data collected by others. Both also offer the temptation to[Page 870] rely on published findings, which means that each method can be subject to publication bias.

However, at least two distinctions exist between secondary analysis and meta-analysis. First, secondary data analysis uses previously collected data to address problems other than those for which the data were collected. In contrast, meta-analysis combines results from multiple studies with similar hypotheses to clarify findings on the same problem. Second, whereas meta-analysis entails using data from multiple studies, secondary data analysis might use data from a single (usually large) study. Because of these differences between the two methods, some researchers have viewed secondary analysis as distinct from all other types of research.

Practical Basis for Secondary Data Analysis

Governments, businesses, schools, and other organizations have gathered much quantitative information. Therefore, when available information gathered for primary research can serve other research purposes, it has made sense to use that information, rather than repeatedly attempting massive data collection efforts. Researchers have highlighted at least four practical reasons for using secondary data.