ABSTRACT
Meta-analysis is a set of statistical methods for combining the results of previous studies. There are differences in the results from study to study. The differences could be caused by one or more of the following types of errors: random sampling errors, random errors of measurement, known systematic errors of which one or more of the researchers was aware, and unknown systematic errors. All four types of errors are possible in any experiment. Hence, the results of any one experiment should be interpreted with caution. However, in this particular example, there are four experiments, which should provide more confidence in any overall conclusions than any single study could. In a traditional, subjective narrative review of these studies, it would be reasonable to synthesize these four experiments with a statement such as "Three of the four experiments on the tutorial instruction showed positive results, with one experiment showing a six-point difference in favor of the experimental group.
