The role of the tutor is important in developing effectivegroup process in educational programs built around small-group,problem-based learning (PBL).The tutor’s roleincludes creating a supportive group climate,encouraging theinvolvement of group members and addressing group prob-lems when they arise.Good tutoring has the potential to enhance group processin both novice and experienced teams.It is worthwhile therefore to monitor the quality of tutorial teaching in PBLprograms and provide valid and useful feedback to individ-ual tutors.In addition to ‘expertise’in teaching,tutors may have con-tent expertise in the area(s) covered by the patient problem(s).Tutors with content expertise may facilitate students’learn-ing,for example,by intervening in students’discussion withtimely statements or questions that evoke relevant ideasand/or clinical reasoning processes.However,studies of theeffect of tutor content expertise on student learning have pro-duced inconclusive results (Schmidt & Moust,2000).Someresearch has found a positive effect on students’achievementin favour of content experts,while other studies have foundmixed or no differences between ‘expert’and ‘non-expert’tutors (see Schmidt & Moust,2000,for a comprehensivereview).The inconclusive results from this research may bedue to differences in (1) the de”nitions of content expertiseand (2) the size of samples used,(3) participating students’level of experience of PBL,(4) students’levels of prior knowl-edge,and/or (5) the amount of structure provided by the medical school curriculum (Schmidt & Moust,2000).The curriculum in Years 1 and 2 of the University ofSydney medical program is structured around PBL tutorials,supported by lectures and other teaching sessions relevant tothe problem.A total of 70 clinical problems are grouped intonine units or ‘blocks’of study.The ”rst block is introductory,seven blocks are based on body systems and the ninth blockis concerned with oncology and palliative care.Prior to eachblock,all PBL tutors are provided with printed materials spe-cially designed to support their teaching for each problem.These support materials,together with extensive web-basedstudent resources linked to the weekly case (readers mayexplore the program web site at http://www.gmp.usyd.edu.au/visitors/),provide a highly structured teaching and learn-ing environment.In this paper,we report the results of an analysis of PBLtutor evaluation data from Years 1 and 2 of the medicalprogram for the period 1998–2000.We conducted a con”r-matory factor analysis on a 24-item tutor feedback form completed by students at the end of each block,and com-pared ratings of ‘expert’and ‘non-expert’tutors,where experttutors were broadly de”ned as having medical training.Weused this de”nition of ‘expert’because tutors with medicaltraining have different levels of clinical experience in additionto knowledge of both the basic and clinical sciences.Non-experts were further subdivided into staff with basic scienceand non-basic science backgrounds.Our analysis does notinclude an examination of the relationship between tutorexpertise and student achievement in the USydMP,becausestudents change their tutors several times during Years 1 and2,and such an examination would involve a higher level ofcomplexity beyond the scope of this article.