In a recent interview I was asked whether students from 'poor' schools should be made lower offers in order to improve their chances of studying at Cambridge or other top Universities.
Four considerations lie behind the reluctance of the Colleges of the University of Cambridge to adopt such an approach:
- The fact that we do already take school performance into consideration as part of our holistic assessment of applicants.
- The ‘ecological fallacy’: students at low performing schools must not be assumed all to be themselves low performing
- Cambridge’s research into the correlation between students’ school achievement and their success or otherwise in our examinations
- The nature of learning and assessment in our degrees, in which timed examinations and an absence of re-sitting opportunities are a central feature.
In assessing applicants we already evaluate their performance in the context of the school(s) and college(s) at which they sat their public examinations, and the year in which they sat them. We gather as much data as possible about each UK applicant’s school, including, where available, its profile of examination results, at GCSE, AS and A Level. A school will be flagged as “lower performing” if its average (capped) GCSE score is 40 or below.1
This definition is derived from extensive research. Among state-sector students those whose school GCSE capped score was below 40 produced stronger GCSE correlations with results in Cambridge than was generally the case, and that the strength of the correlation mounted as school performance declined. The lower the school performance, the more we need to factor in A grades as well as A* in evaluating GCSE results. Nothing in the research suggested that further allowance should routinely be made for weaker GCSE results. We also keep records on how much experience a school or college has in sending students to study at Oxford and Cambridge. This information will be made known to selectors through a system of flags on each application. This gives us a very good sense of the educational context within which a student has achieved his or her results.
In these ways, we do consider carefully the educational context within which results have been achieved. However, the argument that we should consider automatic lower offers for those at schools with lower average scores is based on what seems to be a misapprehension that attendance at a low-performing school equates to being a low performing student and that our expectations of any student from such a school should be lowered. In our experience this is not the case. This is an example of what social scientists refer to as the 'ecological fallacy': the danger in using aggregated or averaged data of assuming that characteristics of the individual must be the same as those that are dominant in a bigger unit, in this case the school.2 If we add to this the consideration that even at 'low-performing' schools a third or a quarter of students achieve five or more A* to C grades at GCSE, we begin to grasp the true nature of the danger in proceeding as though no-one at such a school is capable of the highest results. That they are capable is reflected in the 2012 entry to Cambridge. 63.3% of Home students in this cohort attended maintained sector institutions. When we look at our contextual data flags regarding school performance we find that one in three of these students either attended a low-performing school or a school with a low success rate of admission to Oxford and Cambridge, and something like one in five students had both flags. The hundreds of students we have admitted from the schools at issue form part of a maintained sector entry to the University that has averaged two and a half A* grades each at A Level. These students did not need a lower offer in order to reach Cambridge.
Nor is an argument for lower offers for certain categories of schools supported by the performance of students once here. It is also crucial to stress that at Cambridge students from the maintained sector do not out-perform equally qualified students from independent schools. Both our own research3, and the research most frequently cited by the advocates of lower offers4, confirms this: “differences in achievement by school type largely disappear for students with 30 A level points.” 30 points was the UCAS tariff for AAA at that time (it is now 360 points). This is vitally important, as successful Cambridge applicants typically achieve over 400 points from three A-levels and other qualifications.
Our admissions process seeks to identify and offer places to those students of the highest academic merit with the greatest capacity to thrive on our academically-intensive, fast-paced undergraduate courses. These degree courses are challenging. The pace of work is intensive, courses move ahead rapidly, work is submitted weekly for expert evaluation and discussion in our small-group ‘supervisions’. There are major examinations at the end of each year of study, hardly any provisions for re-sitting5, and little emphasis on coursework in most subjects for most years of study. Excellence in sixth-form examinations is a good indicator of a capacity to thrive in such an environment.
Our research has demonstrated that students with relatively lower attainment in their exams, whether GCSE, AS/A2, IB, or STEP, are at risk of struggling academically in Cambridge. We feel that it would, therefore, be inappropriate to admit systematically a group of students whose results would suggest that they would be at a heightened risk of failure. Such students may be better suited to different modes of teaching, learning, and assessment. This is no reflection on them. It is a reflection on the fit between their aptitudes and the demands of courses here. The risk in admitting them would be that no-one would gain. We do not want to admit students who would find the demands here beyond their capability. We want them to enjoy their undergraduate years, not spend them struggling to survive academically. Admission is only the start: retention is the next step; graduation with a good degree is just as vital a part of this issue.
We have no plans to introduce systematic grade discounts based on school type because our evidence is that there is no justification for such a step.
We take school performance into consideration as part of our detailed appraisal of the academic record of every applicant. It is already a significant consideration in our decision-making. We base our offer levels on research into what will be an appropriate standard to make us confident that a student has every chance of success here. This does not prevent large numbers of students from low-performing schools from reaching Cambridge. They rise triumphantly to the challenge we set them. We do not see a compelling case for reducing our expectations of them. If they meet our offer, they can be sure that they have achieved their place on the back of their academic excellence alone and that we have confidence in their future success.
1 Partington, Chetwynd and Carroll: A Levels are a strong indicator of degree potential.
www.admin.cam.ac.uk/offices/admissions/research/a_levels.html, for a fuller explanation of the methodology, see: www.admin.cam.ac.uk/offices/admissions/handbook/section1/1_4.html
2 I owe this insight to my former colleague, Dr Catherine Sumnall, whose work on contextual data was a valuable contribution to Cambridge's approach.
3 Parks: School background is not a factor in Cambridge degree success, www.admin.cam.ac.uk/offices/admissions/research/school_background.html
4 HEFCE 2003, Schooling effects on Higher Education achievement paragraphs 27, 34 (quoted), 76-78.
5 The exception is for professional qualifying examinations in such subjects as Medicine and even then not for overall re-assessment.