Meta-analysis

In Uncategorized

Meta-analysis is the statistical method by which researchers not only aggregate, but analyze, data from various studies. Writes Saxton (2006) “Meta-analysis is a method for summarizing statistical findings across multiple research studies. It is a useful method for assessing the level of agreement or disagreement surrounding a given research question.” (p.158). However, in order for data from a variety of studies to be analyzed, there needs to be a common methodology in place.

 

Meta-analysis in LIS Research

 

An issue surrounding Library and Information Science (LIS) research is that many researchers do not conform to one known meta-data standard when planning their research methodology. Often, LIS researchers make-up their own parameters when conducting their own research, says Saxton (2006), “… researchers often develop their own operational definition for each new study,” (p. 158). The creation of statistically inconsistent data in these studies results in information that can not be verified in a larger statistical context. Saxton (2006) proposes six basic standards for reporting research, (p. 167), which address everything from including operational definitions for "every variable mentioned in the article," to how to accurately report the significance, or lack thereof, of bivariate relationships.

 

Meta-analysis Resources for LIS Researchers

 

Another resource, Kalyani Ankem’s article, “Approaches to meta-analysis: A guide for LIS researchers,” is an oft-cited example of a good preliminary resource for LIS researchers interested in reaping the benefits of standardized studies. Ankem outlines what researches should consider when designing a study for meta-analysis, and then discusses the three main approaches to meta-analysis, which are the Hedges and Olin approach, the Rosenthal and Rubin approach, and the Hunter and colleagues approach. Bibliographies for these sources are found in the additional resource section.

 

Importance of Meta-analysis

 

Data collected within a consistent and replicable meta-analytic approach can be shared across disciplines for deeper analysis. What does it mean for LIS research that meta-analysis is not often considered when planning a research study? The lack of consistent and standardized data will, writes Saxton, (2006), "... retard the maturation of the discipline by preventing the accumulation of large datasets and enabling new researchers to build upon the foundation laid by experienced researchers. This may also discourage new researchers from pursuing quantitative methods as a possible means of investigation for the questions that interest them," (p. 168). Building upon the work of other researchers allows for a deeper understanding of LIS research, as well as the possibility of a cross-application into other data-driven disciplines, such as science and medicine.

 

Additional Meta-analysis resources

 

Rosenthal, R. (1984). Meta-analytic procedures for social research. California: Sage Publications, Inc.

 

Hedges, L. V., and I. Olkin. (1985) Statistical Methods for Meta-Analysis. New York: Academic Press.

 

Hunter, J.E. & Schmidt F.L. (1990) Methods of meta-analysis: correcting error and bias in research findings. California: Sage Publications, Inc.

 

Bibliography:

 

Ankem, Kalyani. (2005) Approaches to meta-analysis: a guide for LIS researchers. Library & Information Science Research, 27(2), 164-176.

 

Saxton, Matthew L. (2006) Meta-analysis in library and information science: method, history, and recommendations for reporting research. Library Trends, 55(1) 158-170.

 

Sophie Elise Lalazarian