Objective: To determine what variables concerning tutors are associated with the results of the evaluations of their third-year residents.
Design: Crossover observational study.
Setting: All the teaching health centres in the province of Valencia in 1994.
Participants: All the tutors and third-year residents, 78 people in all.
Measurements and results: The information was gathered by original questionnaires, previously validated and capable of contrasting the same information. There was high tutor-resident concordance (0.51-0.65) for teaching activities, supervision of the resident and integration in the Primary Care team (IPCT). There was excellent concordance (0.64-0.92) between residents and tutors for sessions and the resident's research. The variables which differentiated tutors and residents were, according to the ANOVA, Kruskal-Wallis and Chi-squared tests: clinical demand, sessions, number of scientific journals read, number of sessions supervising residents, satisfaction with the resident and IPCT. With the agglomerative iterative method in cluster analysis, four groups of residents were formed, with the significant variables (p < 0.05).
Conclusions: IPCT and clinical activity were the variables which most differentiated tutors from groups of residents.