Sammendrag
Meta-Analysis is a method for combining the results of independent studies. Increasingly, meta-analysis is being used in medicine and other health sciences to augment traditional methods of narrative research by systematically aggregating and quantifying research literature. Recent examples in a medical context are hormone replacement therapy and screening for breast cancer. An example in education is the effect of class size on achievement. A search in Google with the key words "meta-analysis" and a substantive field or topic such as "gender differences" or "math education" will uncover numerous examples of the use of meta-analysis in the weighing of evidence. In this talk I will discuss the origins of meta-analysis, reasons for its popularity, as well as some of issues such as bias, that have ensued. I will also present some of the statistical methods used. Nonparametric methods have now gained prominence, especially in genetic applications in which the data consists of a very large matrix. The most common methods in the social sciences are based on regression models, and in the medical sciences they are based on the binomial or multinomial distribution.