A fairer deal for top International Baccalaureate students

Francis Green
It seems that Leeds University and Kings College London have decided to become more generous to their applicants with the International Baccalaureate, lowering the grade equivalents with A-levels. Where once they asked for 39 points (out of a rare maximum of 45), now they will ask for around 35. A good thing too. For too long now, the upper tier of universities have been far too snooty about the IB. Our research at the IOE showed that top universities did not appreciate how well their IB students were doing.
We compared IB and A-level students who make the same decision on university and subject. The principle we used is that the equivalence map between the IB grades and A-levels should be such that IB and A-level students on average are shown to do equally well. We deployed the power of large numbers. Looking at all those students who graduated in 2010, we compared our equivalence map with the map that universities were in practice using. We found that:

• In the middle ranks the universities’ guesses were broadly right: IB students performed similarly to A level students in middle-ranking universities that recruited IB applicants with grades in the low 30s. At the bottom end, students who barely passed the IB (pass mark 24) performed a little worse than their fellow students who had been admitted with low A-levels.
• At the top end of the scale, however, universities’ IB students performed significantly better than those they mistakenly thought were similar A level students. In other words, the ‘high-ranking’ universities were asking for too high grades from their IB applicants. Any IB students rejected because they did not quite achieve the very high grades asked for (around 38, 39) could feel aggrieved, as they would have done at least as well on average at their chosen subject as accepted A level students.
• We excluded Oxford and Cambridge from our analysis because of their intensive selection processes which dominate exam grades as their selection filter. Nevertheless, their admissions tutors frequently appear to ask for unreasonably high IB points (often as high as 42 at Cambridge) compared to the grades asked of their A level students, most of whom have very little trouble in achieving the top A-level grades asked for, once given their conditional offers.

Now don’t get me wrong: it is quite easy to misread research of this nature, thinking that it proves IB is a better preparation for university than A-levels: it does not. As anyone with any contact with IB knows, on the whole the IB has attracted more academically capable students over the years. Of course, more academically-capable students do better academically! The argument that IB is best might be valid, but it would be difficult to set up a statistical test, not to mention a quasi-experiment, that would seriously evaluate that claim. It is very difficult to find adequate ‘controls’ that mimic the counterfactual case for a school-student that opts for A-levels instead of IB.