• I C McManus.
• Centre for Health Informatics and Multi-professional Education, Royal Free and University College Medical School, University College London, Whittington Hospital Campus, London N19 5NF. i.mcmanus@ucl.ac.uk
• BMJ. 1998 Oct 24; 317 (7166): 1111-6; discussion 1116-7.

ObjectiveTo assess the relation between a range of measures and the likelihood of applicants to medical schools in the United Kingdom being offered a place overall and at each medical school, with particular emphasis on ethnic minority applicants.DesignData provided by the Universities and Colleges Admissions Service on 92 676 applications to medical schools from 18 943 candidates for admission in 1996 and 1997. Statistical analysis was by multiple logistic regression.Main Outcome MeasuresReceipt of a conditional or unconditional offer of a place at medical school.ResultsEighteen separate measures were independently associated with the overall likelihood of receiving an offer. Applicants from ethnic minority groups were disadvantaged, as were male applicants, applicants applying late in the selection season, applicants making non-medical (so called insurance) choices, applicants requesting deferred entry (so called gap year), and applicants at further or higher education or sixth form colleges. Analysis at individual medical schools showed different patterns of measures that predicted offers. Not all schools disadvantaged applicants from ethnic minority groups and the effect was stable across the two years, suggesting structural differences in the process of selection. The degree of disadvantage did not relate to the proportion of applicants from ethnic minority groups.ConclusionsThe data released by the Council of Heads of Medical Schools allow a detailed analysis of the selection process at individual medical schools. The results suggest several areas in which some candidates are disadvantaged, in particular those from ethnic minority groups. Similar data in the future will allow monitoring of changes in selection processes.

16 keywords...

hide…