Even one star at A level could be "too little, too late" for medical student selection

Abstract

Background: More and more medical school applicants in England and Wales are gaining the maximum grade at A level of AAA, and the UK Government has now agreed to pilot the introduction of a new A* grade. This study assessed the likely utility of additional grades of A* or of A**.

Methods: Statistical analysis of university selection data collected by the Universities and Colleges Admissions Service (UCAS), consisting of data from 1,484,650 applicants to UCAS for the years 2003, 2004 and 2005, of whom 23,628 were medical school applicants, and of these 14,510 were medical school entrants from the UK, aged under 21, and with three or four A level results. The main outcome measure was the number of points scored by applicants in their best three A level subjects.

Results: Censored normal distributions showed a good fit to the data using maximum likelihood modelling. If it were the case that A* grades had already been introduced, then at present about 11% of medical school applicants and 18% of entrants would achieve the maximum score of 3 A*s. Projections for the years 2010, 2015 and 2020 suggest that about 26%, 35% and 46% of medical school entrants would have 3 A* grades.

Conclusion: Although A* grades at A level will help in medical student selection, within a decade, a third of medical students will gain maximum grades. While revising the A level system there is a strong argument, as proposed in the Tomlinson Report, for introducing an A** grade.

Figure 1

Solid black bars show observed distribution of A level points from best three A level grades, scored as A = 10, B = 18, C = 6, D = 4, E = 2 for a) all UCAS applicants, b) all medical school applicants and c) all medical school entrants. Yellow bars show fitted results of a censored normal distribution. The yellow bar at 30 points (AAA) can be decomposed into the red bars for the fitted, uncensored normal distribution, scored as 12 = A* and 14 = A**. The ordinates of medical school applicants and entrants are on the same scale to allow comparison.