You are here: Home / Types of research evidence / How are the effects of teaching on learning to be measured?

How are the effects of teaching on learning to be measured?

Prepared by John Church, PhD, School of Educational Studies and Human Development

University of Canterbury, Christchurch, New Zealand.

Once the learning researcher or teaching researcher has developed an accurate and reproducible measure of learning it becomes possible to seek the answer to many interesting questions. For example learning researchers may seek the answer to questions such as “What are the conditions which must be provided in order for students with achievement level “z” to learn and remember how to do “y”? Teaching researchers may seek the answer to questions such as “What is the effect of teaching procedure “a” on the rate of acquisition of skill “b” in children with achievement level “c”?

This chapter examines the kinds of research methods which can be used to identify relationships between teaching events (and other kinds of events) and student learning. We pay particular attention to the kinds of research methods which can be used to identify cause-and-effect relationships because these are the ones which are of greatest interest to teachers. The chapter is divided into the following sections.

Section 1 examines the kinds of research questions which are possible and groups them into a number of categories according to the research method required in order to answer the question.

Section 2 focuses on methods which can be used to identify functional relationships (that is, cause and effect relationships) and examines the relationship detection methods which have been developed by behaviour analysts, cognitive and social scientists, and ethnographers.

Section 3 evaluates the relative utility and productivity of the experimental relationship detection procedures developed behaviour analysts, cognitive scientists and ethnographers.

Section 4 discusses the quality characteristics which must be met if an experimental analysis is to yield a believable result. The following experimental design features are identified and discussed: the reliability (reproducibility) of the experimental procedure, the accuracy of the measure of experimental effects, the validity of the conclusions drawn from the experimental result, and the chain of inference which is involved in interpreting the result of an experiment in learning (or teaching).

Section 5 examines the accuracy of the inferences which are being drawn from the results of (a) the within-subject experiments of the behaviour analyst and (b) the randomised groups experiments of the cognitive scientist and identifies some of the problems arising from the way in which significance testing procedures are currently being employed.