Improving the Assessment of Teaching Effectiveness With the Nonequivalent Dependent Variables Approach

Abstract

Introduction: Assessing teaching effectiveness is relevant for improving one’s teaching and for moving through the tenure process; however, the validity of assessment methods, such as Student Evaluations of Teaching (SET), have been heavily criticized. Statement of the Problem: Using a one–group pretest–posttest design and assessing learning over the semester has several advantages over SET; however, one drawback is in making conclusions about the cause of changes in the post-test. A change could be due to learning in the semester, maturation, history, or even a testing effect. Literature Review: To improve the inferential quality of teaching assessment, a nonequivalent dependent variable (DV) design is highly advantageous. A nonequivalent DV is an outcome that is not the target of the intervention yet responds to the same contextually relevant factors. Teaching Implications: By using a nonequivalent DV design, there might be an increase from the beginning of the semester to the end of the semester in the main DV, but no increase in the nonequivalent DV, which provides a stronger argument that the change in the main DV is due to a true learning effect. Conclusion: Using nonequivalent DV methodology improves inferential quality and is easily implemented.