ABSTRACT
Meta-analysis is a set of statistical methods for combining the results of previous studies.1 To understand its basic premise, consider the following example. Suppose a researcher (named W) randomly assigned 50 students to an experimental group that received one-on-one remedial tutorial reading instruction. The remaining 50 children were the controls. At the end of the experiment, the experimental group had a mean of 22.00, and the control group had a mean of 19.00 on a standardized reading test. Subsequently, three other experimenters conducted strict replications of the first experiment using the same research methods, the same type of tutorial reading instruction, and the same number of second grade students drawn at random from the same pool of students (e.g., second-graders in a particular school district). The posttest means of the four experiments are shown in Table 1. Table 1 Results of a Meta-Analysis of Two Experiments
In Table 1, 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: 1. Random sampling errors created by assigning
participants at random to the two groups, such as having more motivated students assigned (quite at random) to the experimental group in one or more of the experiments.
