Using Nonequivalent Dependent Variables to Reduce Internal Validity Threats: Rationale, History, and Examples from Practice




Watch the video

View the slides

Legitimate knowledge claims about causal relationships have been a central concern among evaluators and applied researchers for several decades and often have been the subject of heated debates. Since publication of Campbell and Stanley’s (1966) Experimental and Quasi-Experimental Designs for Research, which was followed more than a decade later by Cook and Campbell’s (1979) Quasi-Experimentation: Design and Analysis for Field Settings, and thereafter by Shadish, Cook, and Campbell’s (2002) Experimental and Quasi-Experimental Designs for Generalized Causal Inference, alternative explanations for effects or outcomes observed from studies of applied interventions have been the bane of practicing social scientists and evaluators. Collectively, these alternative explanations are generally known as threats to validity and are largely related to the correctness of or support for one or more types of inference that arise from investigations of applied interventions. In this Evaluation Café the rationale for, history of, and examples from practice of using nonequivalent dependent variables as one strategy to reduce the number and plausibility of certain types of internal validity threats will be presented and discussed.

No Comments

No comments yet.

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.