UQ admission irregularity

Recently it has been in the news that a relative of Vice Chancellor Paul Greenfield was accepted into a program without satisfying the same requirements of other applicants. On the 9th of December, UQ released a memorandum detailing the offers of undergraduate-entry MBBS applicants (which I read from Paging Dr), and it included talk of the irregularity. http://www.uq.edu.au/about/planning-docs/2011-mbbs-program-admissions/

Regarding the irregularity itself, it seems like the Vice Chancellor and his deputy have both paid the price and stood down. Then there is debate on whether the relative who was given the irregular offer should keep their place or not. Personally, I don't have a very strong view on this.

In that memorandum, there was also noticeably a mention of people with Vice Chancellor scholarships (by achieving an ATAR of 99.95; the highest possible rank) who did not do well on the UMAT but still got irregular offers. This has never been the official policy of UQ to give such offers in such circumstances, but it should be noted that this is very similar to what is done in Melbourne (needing 99.90 and above) and USyd (needing 99.95). The main difference between Melb and USyd compared to UQ is just that UQ didn't interview for this, but UQ doesn't interview anyone except ATSI applicants anyway. It does raise eyebrows when UQ does things not in their official policy though, and it seems like they won't be doing such a thing next year. I think they could be justified if they made giving medicine places to people with Vice Chancellor scholarships regardless of UMAT official policy though.

This raises another issue. Is using UMAT a good system to select applicants? I personally do not think so, and so far the evidence seems to suggest that UMAT doesn't predict medical school performance well at all, although others have criticized the study because it possibly has "ceiling effects" there. I don't think that UMAT is any better than still using academic achievement (ATAR) even at the high levels, although I have some issues with English being compulsory primary 4 in ATAR calculations in Victoria and NSW (but not some other states; in some it is not even compulsory in year 12), without Math being compulsory at all in some states (including VIC and NSW), let alone primary 4.

Similar criticisms are sometimes applied to the GAMSAT, and I don't think the GAMSAT is as optimal as it can be either, however it does test science unlike UMAT, and that section has some modest predictive value. The MCAT in North America seems to have more solid evidence behind it though, so perhaps if we modify the GAMSAT to be more of that style then it can probably predict medical school performance better. However, it will probably also disadvantage non-science background students, so if unis here think that that type of diversity is important, then such a change in the test probably can't happen. I think a standardized test like the MCAT or GAMSAT with some science is justified in selecting med school applicants for graduate entry, especially because for GPA there hardly any scaling for differentiating between subjects which are easier or harder to score highly in, between different unis and within them.

On the other hand, for ATAR, at least in most states, there is scaling to take into account differences in difficulty between subjects, and in most cases standardized exams for most of the marks going towards a subject. So there is less need for a standardized test including a science section; it would seem to just duplicate whatever assessment is done at school, especially if chemistry is a prerequisite (as it is in some unis for medicine including Monash but not all. Funnily, biology is not a required prerequisite anywhere in Australia for undergraduate entry, and I didn't do year 12 level bio in high school, but that's a discussion for another day).

It has been argued that ATAR discriminates against people of lower socioeconomic status. This may be true, but it is also true that UMAT preparation courses are expensive, and it is hard for people of low socioeconomic status to afford those too. I think if we want to get people of low socioeconomic status into medicine and other uni courses, then we should have some sort of affirmative action instead of using some dubious selection tools which may add more noise into the selection process and being not very useful predictors.