Chapter two - Research Design and Issues of Validity

Summary

The goals of empirical research in social psychology can be differentiated into three broad categories: demonstration, causation, and explanation. Research performed for the purpose of demonstration is conducted to establish empirically the existence of a phenomenon or relationship. The tripartite distinction between internal, external, and construct validity, provides the basis for organizing the discussion of validity issues. Construct validity refers to inferences made at both stages of research linking concepts to operations. The chapter uses the concept of demand characteristics to illustrate the difference between methodological confounds (which affect construct validity) and methodological artifacts. Construct validity represents one form of generalizing from the observed results of an empirical study to conclusions that go beyond the results themselves. The debate includes discussions of whether there are necessary trade-offs among the various aspects of validity or whether it is possible to demand that research maximize internal, external, and construct validity simultaneously.