Size matters: what influences medical students' choice of study site?

BACKGROUND: The University of British Columbia, Canada doubled its class intake in 2004, creating, in addition to its main metropolitan campus, 2 distributed campuses, one in a medium sized island city, and the other in a small geographically isolated northern city. Our admission process attempts to identify students more suitable for education in our northern, rural site. Students also indicate their preferred site. Little is known about what influences student choice when they have more than one campus to choose from at a single medical school. AIM: To understand what influences students' preference of study site in a single medical school with 3 separate campuses, one with a rural mission. METHODS: We used qualitative methodology to examine what influenced student choice of study site. Semi-structured interviews were conducted with students at all three sites (n = 37). Iterative and independent coding and analysis took place to corroborate research findings. RESULTS: The primary theme was size of class and community. Some students viewed a larger class size and larger study and practice community as advantages, others viewed a smaller class size and smaller study and practice community as important. Additional themes were perceptions of quality of education, relationships, and lifestyle. These were related to the larger theme of class and community size and overlapped. Students articulated advantages and disadvantages of each site, and dynamic tensions in their choice of sites. Close relationships and educational experiences were emphasized at the smaller regional sites. Greater access to medical and educational specialists and the diversity offered by a larger, more anonymous class, patient case-mix, and community were emphasized at the originating and largest site. Partner and family--trumps--could overrule preferred site choices. CONCLUSION: Early and comprehensive descriptions of the differences between sites for students and their partners is needed to help truly informed choices.

Medical student selection: choice of a semi-structured panel interview or an unstructured one-on-one interview.

Reliability has been shown to be higher in structured medical admissions interviews as compared to unstructured interviews. This study reports the comparison of a proposed semi-structured panel interview with a current individual unstructured medical admissions interview. Inter-rater reliability coefficients were calculated, and correlations were estimated between panel, individual and academic scores. Admission status in 2003 was related to these scores by means of logistic regression. Both individual and panel interviews were significantly correlated with admissions status. The inter rater reliability coefficient (from individual interviews) was 0.12 whereas the interpanel reliability coefficient was 0.52. Panel interview: good across panel and within panel consistency of scoring. No effect of who asked the questions, question order, or interview duration on scoring. No correlation between panel interview scores and academic variables (MCAT, GPA). We found good inter-panel reliability, a high consistency within and between interview panels, and uniformly positive questionnaire responses. The panel interview measures something different from academic variables. These data, in conjunction with a strong sense from the medical and psychological literature supporting the reliability and validity of a semi-structured panel interview, support our decision to replace our individual interview with the panel interview.