A 50-item observational instrument was used by cooperating teachers and college supervisors to evaluate the competencies of 77 home economics student teachers from two universities at two time intervals during the student teaching experience. Two analyses of variance (AOV) for each of the 50 items were used to identify items which judges could rate reliably. Intra-class correlation coefficients were utilized to ascertain if different ratings between student teachers were due to differences in their performance. The AOV analyses suggested that judges could reliably rate 18 items. Of these, five appeared to discriminate between student teacher's performance whereas eight did not. (Author)