Skip to main content

Beware the social engineers

Tinkering with university admissions is not the way to eradicate bias, says Geoff Lucas

Consultation on the first report by Professor Stephen Schwartz's committee, which is looking at ways to create a fairer admissions system for higher education, is now in full swing. Set up by the Government after last year's controversy over Bristol university's admissions, its task was never going to be easy. As the report acknowledges, words like "merit" and "potential" are susceptible to a range of interpretations, many of them highly value-laden.

There are many aspects of the report one could challenge. For example, its over-reliance on American models which are inappropriate to the English context. Or its failure to acknowledge that the identification of potential at age 18, as opposed to 11 or five, is too little too late.

Press coverage to date, however, has highlighted two of the more emotive issues.

* Should certain students be offered places with lower examination grades than others?

* Should the type of school attended by an applicant be the basis for such differential offers?

While the media tends to simplify the issue (maintained versus independent schools), the report itself opens up a more subtle differentiation: high-performing versus low-performing schools. Both are relative terms, but an applicant who rises above the average for his or her school (in terms of exam performance) is, the report argues, more likely to do well at university than someone with the same grades which are only average in another school.

In both cases the report cites research to justify such proposals, and the questions in the consultation document are clearly framed to marshal support for them.

Yet elsewhere in the report, there is abundant (and more robust) evidence which should, if looked at dispassionately, point to a very different set of policy proposals.

Here is a selection of statements, taken verbatim from the report, followed by alternative consultation questions which could have been put.

1 Across all subjects and HE institutions, there is no objective evidence of bias either for or against students from particular schools and colleges or from particular socio-economic groups. Question: is the Schwartz exercise needed at all?

2 It is differential rates of application rather than bias in admissions procedures that is the main cause of under-representation of disadvantaged groups in top universities. Question: shouldn't applications, rather than admissions, be the focus of any review?

3 The principal cause of the differential participation in HE by social class is the big discrepancy in pevious educational attainment.

Question: isn't the improvement of primary and secondary education the real key to widening access and participation at tertiary level?

4 Previous attainment, as measured by A-level grades, or equivalent, is the best single indicator of academic success and retention at undergraduate level.

Question: how will attempts to marginalise A-levels help to improve standards in HE?

5 Nine out of 10 students who obtain two or more A-levels currently enter HE by age 21; this progression rate is fairly even over all social classes.

Question: shouldn't the Government be trying to get more young people to achieve two or more A-levels, instead of tampering with admissions?

6 Of those who obtain 25 or more A-level points by the age of 18, 97 per cent from non-manual socio-economic groups go on to HE, as do 94 per cent from the three manual socio-economic groups.

Question: as for Question 5 but applied to high-achieving students who are most likely to apply to the most selective universities.

7 Offering lower grades to students in maintained schools:

* might not work in all subjects at all universities (where specific knowledge or skills are required); * could unfairly disadvantage students in independent schools; * is unlikely to be of much help at the upper end of the performance scale where "school type" effect is weakest.

Question: if the blanket application of contrived, statistical correction factors is both unfair and ineffective (not to mention patronising), why even consider such an option?

I offer one final example (also from the report itself) of the need for caution before trying to engineer specific outcomes in crude quantitative terms. The evidence suggests that "people from ethnic-minority groups are over-represented in HE in England. They accounted for 14 per cent of full-time and part-time HE students in 20012, compared with 9 per cent of 14 to 18-year-olds in the population as a whole".

Would anyone in their right mind recommend, on the basis of this evidence, that students from other groups should now be set lower grades in order to compensate? I very much doubt it.

The current system of university admissions may not be perfect, but the evidence provided by the Government's own independent committee does not appear to justify some of the options presented by that same committee.

Access procedures overseen by professionals, warts and all, must be infinitely preferable to formulaic rules determined and imposed by politicians.

Geoff Lucas is secretary of the Headmasters' and Headmistresses' Conference

Log in or register for FREE to continue reading.

It only takes a moment and you'll get access to more news, plus courses, jobs and teaching resources tailored to you