The Retrospective Pretest in an Educational Training Context

The present study investigated the relative validity of the conventional pretest-posttest design and the retrospective pretest-posttest design using self-report measures as outcome criteria. As predicted the experimental training in Seeing Problems Strategy was effective in improving subjects’ task performance. Traditional self-report pre-post comparison did not reflect this actual change of performance, whereas the retrospective self-report approach did. Comparison of pretest mean scores and retrospective pretest mean scores indicated that subjects initially over-estimated their performance. This result represents further evidence on the validity of the response-shift notion as introduced by Howard et al. (8) and supports earlier research in showing that the retrospective pretest is a valid means to control for this source of contamination.