A component analysis of a teacher training program

The present study is a component analysis of a comprehensive teacher training program. The study investigated the effects of three types of training on graduate teaching assistants' (TAs) teaching behavior, perception of teaching ability, and ratings of effectiveness by their students. Thirty-eight TAs either (1) participated in a comprehensive training program, (2) received videotape feedback and consulting, (3) received mid-term student ratings feedback and consulting, or (4) no training. Analysis of the data using multivariate analysis of variance indicated that the treatments were effective in producing positive changes in teachers' self-ratings of teaching ability, teaching behavior, and marginally in student rated progress relative to a no-treatment control group. A priori comparisons denoted that the full treatment and videotape groups changed significantly more than the other groups but were not significantly different from each other. In addition, retrospective pre tests were collected to ascertain possible response-shift bias. Comparisons of pre/post versus retrospective pre/post self-reported change scores confirmed the hypothesis that response-shift bias was operating and that retrospective pre-tests were a more valid index of self-reported change. Implications and methodological considerations for future teacher training programs are discussed.