PURPOSE: To identify aspects that influence students' evaluation of the overall quality of clerkships and learning in clinical settings. METHOD: The authors analyzed 2,450 questionnaires dated 1997 through 2005 that evaluated clerkships of seven medical specialties (internal medicine, surgery, pediatrics, psychiatry, community medicine, emergency medicine, and obstetrics-gynecology). Students rated 22 questionnaire items addressing clerkships' global evaluation and domains related to structure, supervision, and clinical and problem-solving learning (PSL) activities using a five-point Likert scale. The authors performed statistical analysis using principal component analysis and regression analysis of items associated with students' global evaluation of clerkships. RESULTS: Correlation between clerkships' global ratings and ratings derived from the evaluation questionnaire was 0.871 (P <.0001). Clerkships' quality was mainly related to their organization, students' integration into clerkship, improvement of clinical skills, supervision, and residents' availability (r = 0.405; P <.0001). Among learning activities, opportunities for clinical practice predominated as the contributing factor to the overall perceived quality of most clerkships, but less than PSL activities in psychiatry (r = 0.070 versus 0.261, respectively; P <.001) and community medicine (r = 0.126 versus 0.298, respectively; P <.001); in surgery, both clinical practice and PSL activities contributed minimally to the clerkships' perceived quality (r = 0.150 and 0.148, respectively; P >.05). CONCLUSIONS: Factors influencing students' evaluation of a clerkship vary among medical specialties and depend not only on the teaching and teacher but also on the clerkship's organization, supervision, and learning activities. For clerkships where direct and multiple access to patients is more difficult, written case-based PSL activities proved complementary to direct patient encounter activities.