ABSTRACT
This chapter is about experiments and what can be learned from them about standards supporting valid cause-and-effect conclusions. How can an experiment control for all variables that might account for an observed relationship between the treatment variable and the outcome? The short answer is, by random assignment. After the treatment, the difference between the treatment and control group average for the dependent variable can’t be due to anything other than treatment effects and random variation. Social experiments using random assignment with people are not usually feasible, due to ethical and other reasons, as illustrated in this chapter. However, the study of experiments provides important insights about how to evaluate non-experimental research, as well as how to evaluate information from everyday life. Random selection should not be confused with random assignment. Random selection is for “external” validity of the experiment: Does the observed effect generalize to populations other than the one from which subjects were chosen? Five basic research designs are examined, along with everyday and other examples for each. The discussion in this chapter is intended to help you develop insights and skills to become a better judge—in your future career and everyday life—about what is and is not likely to be true.
