Classroom Management That Works: Research-Based Strategies for Every Teacher

The concept of a meta­analysis was briefly addressed in Chapter 1. Here we consider this form of research in more depth, with particular attention to the meta­analysis that forms the basis for the recommendations in this book. (For a more technical description of the meta­analysis conducted during the development of this book, see Marzano [2003a].)

In very general terms, a meta­analysis of the research on any topic attempts to summarize the findings from the studies on that topic in quantitative terms. Typically, the findings from the studies reviewed are all translated into an effect size, which is defined here in the following way:

Click for a larger version

By definition, this formula translates the difference between the mean of the experimental group and the mean of the control group into standard deviation units. In the context of this book, the experimental group was typically a class that employed some type of classroom management technique, and the control group was a class that did not. In Chapter 1, I described an effect size in terms of the difference in average number of disruptions between the experimental group and the control group. However, one confounding factor in terms of interpreting the findings of the meta­analysis conducted for this book is that some studies looked at one individual and others looked at whole classes. An effect size is interpreted slightly differently when a study is based on an individual as opposed to experimental and control groups. To illustrate, with single­subject designs, the misbehavior of an individual student was typically tracked before and after a particular classroom management technique was used. A technical explanation of how these pre­intervention

-116-