Using a Retrospective Pretest to Reveal the Effects of a Teacher-Leadership Development Project
Abstract
"We evaluated a math teacher-leadership development project using a self-report pretest-posttest design including retrospective pretest ratings to control for response-shift bias. Math teachers from Cohort 1 took the pretest, posttest and gave retrospective pretest ratings. Pretest-posttest comparisons indicated that the project was not generally effective. However, a comparison of retrospective pretest ratings with posttest ratings indicated that the project significantly increased leadership development generally. Similar results were obtained when we used self-efficacy for teacher leadership as the dependent variable. This suggests that project effects were masked when only pretest-posttest comparisons were made. Self-efficacy change was the only predictor of increase in teacher-leadership activities. We outline changes in research and project design made for Cohort 2 based on these findings."
