Influence of objective measures on self-reports in a retrospective pretest-posttest design.

Conducted 2 experiments to assess the influence of awareness of an objective measure and completion of a performance test on the self-ratings of 152 Dutch undergraduates. Both studies involved a pre-/posttest design, including retrospective preratings. Results indicate that the retrospective prerating approach reflected actual improvement of performance more accurately than the conventional pre/post comparisons. The response shift phenomenon--defined as a significant difference between mean preratings and mean retrospective ratings--was demonstrated in both experiments. It was observed that, while awareness of an objective measure tended to weaken the response shift phenomenon, actually taking a performance test tended to strengthen it.