Causal determinants of response shift bias

Journal Title
Journal ISSN
Volume Title

In the evaluation of interventions employing traditional pretest-posttest self-report data, response shift bias has been shown to jeopardize internal validity. The use of a retrospective pretest-posttest design has been shown to eliminate response shift bias. The present study contrasts both designs in examining helping skills training outcome. In addition, a replication of the Trainability Index (Anthony, Gormally, and Miller, 1974) as an effective predictor of potential benefit from helping skills training is examined. The Trainability Index is evaluated in regard to the possible influence of a personality factor; namely, self monitoring (Snyder, 1974). Results of the study showed that the helping skills training program was effective in increasing empathy functioning skills. However, nene of the proposed hypotheses were supported. Problems with the behavioral measure and limited sample size were viewed contaminating factors.

Counseling--Ability testing, Psychological tests