Paperonium

Medical admissions · BMJ · 2012

For non-professional-family applicants at some schools, the summary prints another group’s range: 0.45–0.60, not 0.32–0.81.

The summary:non-professional familiestight rangeThe table: another group,other schoolssame rangeThe table:non-professional families3 times as wide0.45 to 0.60×0.32 to 0.81×0.25×0.4×0.6×0.8×1×same odds as other applicants
The summary:non-professionalfamiliestight rangeThe table:another group,other schoolssame rangeThe table:non-professionalfamilies3 times as wide0.45 to 0.60×0.32 to 0.81×0.3×0.5×0.7×1×same odds asother applicants

Times other applicants’ odds of being offered a place. Dot: figure; bar: range.

The paper says

The summary says that at UK medical schools using an admissions test, the UKCAT, only for close calls, applicants from non-professional families (the two lowest of five social classes) had 0.51 times others’ odds of being offered a place: about half, range 0.45 to 0.60.

Its own data show

Its own table gives the same 0.51 but a range of 0.32 to 0.81, about three times as wide, which matches the table’s other figures for this group. The summary’s 0.45 to 0.60 is exactly the table’s range for another group: pupils from non-selective, non-fee-paying state schools, at medical schools that used the test differently.

A range shows where the true figure very likely lies: the narrower, the more precisely it is known. Odds are a measure of chance; 1 times others’ odds means equal chances. Both ranges stay below 1, so the disadvantage holds, but the data place it anywhere from a third to four-fifths of others’ odds, not the summary’s tight band.

“The summary” means the paper’s abstract: the short summary printed at its top, which is the part most readers see.

Read the paper →Full technical write-up →

Widening access to UK medical education for under-represented socioeconomic groups: modelling the impact of the UKCAT in the 2009 cohort · BMJ (2012) · doi:10.1136/bmj.e1805