Skip to Main Content


Research designs are intended to provide structure that allows for logical conclusions about the relationship between independent and dependent variables. The investigator must have confidence that extraneous factors have not influenced the outcome. Even with a design that fulfills the requirements of an experiment, such as a randomized controlled trial (RCT), researchers must be vigilant regarding many potential sources of confounding that can obscure the effects of an intervention. Confounders may be extrinsic factors that emerge from the environment and the research situation, or they may be intrinsic factors that represent personal characteristics of the subjects of the study.

In Chapter 10, we addressed the importance of validity for measurements to have confidence in the meaning of a measured outcome. Here, we are concerned about validity in relation to the design of a research study and interpretation of results. The purpose of this chapter is to examine issues of control that must be addressed in the design and analysis of research. Although these concerns will be presented in the context of explanatory studies, they are also relevant to quasi-experimental and observational designs.

Validity Questions

Regardless of the care we take in the design of research, we know that clinical studies seldom have the ability to completely eliminate confounding effects. Although causality can never be demonstrated with complete certainty, the experimental method provides the most convincing evidence of the effect of one variable on another.

The goals of explanatory research correspond to four types of design validity (see Fig. 15-1). These form a framework for evaluating experiments: statistical conclusion validity, internal validity, construct validity, and external validity (see Table 15-1).1

Figure 15–1

Four types of design validity. Each form is cumulatively dependent on the components below it.

Table 15-1Threats to Design Validity

Statistical Conclusion Validity

Is there a relationship between the independent and dependent variables?

Statistical conclusion validity concerns the appropriate use of statistical procedures for analyzing data, leading to conclusions about the relationship between independent and dependent variables. Some specific threats to statistical ...

Pop-up div Successfully Displayed

This div only appears when the trigger link is hovered over. Otherwise it is hidden from view.