14. Meta-Analysis
Meta analysis will be used to combine results of similar studies quantitatively to give an overall picture of the problem undertaken. One has to have explicit criteria for inclusion to avoid combining studies that have fundamental differences in design. The strengths of a meta-analysis is that it functions essentially like a mega-study with increase in sample size and greater generalizability since data is being captured from studies conducted at multiple sites, by differing groups of investigators.
To start a meta analysis we will have to identify the articles available with respect to the problem we have undertaken and then combine the articles which meet our pre-determined inclusion criteria.
Recommended Resources
Chatfield C, Collin A. J. Introduction to multivariate analysis. First edition, Chapman and Hall, London; 1980.
Flury, B, Riedwyl, H. Multivariate statistics: A practical approach. Chapman and Hall 1988.
Hyperstat Online at: http://davidmlane.com/hyperstat/Statistical_analyses.html.
Sackett DL, Haynes RB, Guyatt GH, Tugwell P. Clinical Epidemiology: A Basic Science for Clinical Medicine. 2nd ed. Boston, Mass: Little Brown & Co Inc; 1991:145-148.