Response Shift Bias in Pre- and Post-test Studies.

Sir,

This is in regard to the article, “Role of clinical images based Teaching as a Supplement to Conventional Clinical Teaching in Dermatology” published in Indian J Dermatol 2015;60(6):556-61.[1]

The authors did a commendable job to assess the effect of clinical images-based teaching as a supplement to patient-based clinical teaching in dermatology, among final year MBBS students.

In continuation with the current article, I want to emphasize on an important aspect of evaluating knowledge with self report pre- and post-test measures versus then post test measures.

In the current study, students were asked to fill in a questionnaire “after the class,” to rate their knowledge and skill before class and after class. Student self-report measures of change are widely used in evaluation research to measure the impact and outcomes of an educational program or intervention.

Conventionally to evaluate the impact of an intervention (administering clinical images as a method of teaching in the present study) on change in student's level of knowledge and skills, a comparison of the student's pretest scores with their posttest scores was made. However, that method of evaluating change came out to be problematic due to the confounding factor of “response shift bias.”[2]

The alternate is to use the “retrospective post-then-pre” design. This differs from the traditional pre- and post-design in that both before and after information is collected at the same time, i.e., after the intervention.

“Response shift bias” phenomenon is a source of contamination of self-report measures that result in inaccurate pretest ratings.[3] In the traditional “pre-post” design, learners answer questions before an educational program, engage in the lesson, activity, or course, and then answer the same questions again after finishing the program. In the “retrospective post-then-pre” design, both before and after information is collected at the same time, i.e., after intervention whatsoever. The difference between the “pre-post design” and “retrospective post-then-pre design” (in the “pre” component) is referred to as response shift.[3]

Hence in conducting such education interventional studies, it is recommended that researchers collect “retrospective post-then-pre” data along with the traditional “pre-post self-ratings” and with any other objective measures if available which will help provide a more complete assessment of the effect of the intervention on the study group.[3]

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.