Bac is way forward

Bill Stephenson offers some alternatives to an unfair system of gauging a student's potential

Admissions policies have been much in the news lately. As the A-level results become known this week, many admissions tutors must be asking themselves whether A levels on their own are really the best or fairest method of selecting students.

Some argue that all students should be selected after their A-level results are known. This would be a terrible mistake and would lead to the tyranny of the A-level score over everything else. On the other hand, there are those who argue that A-level grades are no great shakes at predicting the class of degree a student will achieve and so their validity as a method of selection is dubious.

It has certainly been my experience that A-level grades, while not perfect, are not bad predictors of a student's ability to cope with and benefit from a particular degree. Until they are replaced by something better, they will have to remain one of the main weapons in the admissions tutor's armoury. I would rate some sort of system based on the international baccalaureate as the way forward.

Private schools with their better resources can often obtain an extra A-level grade for a student than a state school. How can we compensate for this? How should we discover and assess potential? One way is to ensure that all students who have the potential to complete and benefit from a particular degree should be interviewed. But this is an extremely time-consuming and labour-intensive process, and many universities have abandoned the individual interview for "open days".

In some very popular subjects, such as law and medicine, admissions tutors will be overwhelmed by well-qualified applicants and so will have to choose whom to call for interview. This is often done solely on the basis of the Universities and Colleges Admissions Service form, whose quality can depend on the ability of the student's school to play the system. It is incumbent on university departments of law and medicine to be able to justify their decisions in a transparent and objective way. Many departments ask for extra information, such as samples of work, or set the students a piece of work to complete.

Those departments that do interview should do so well. Their staff should have received training and should have been set clear objectives and guidelines. We all know that some students may have an "off" day. In most cases, even after a favourable interview, most admissions tutors will still want clear evidence of a student's ability, and so a conditional offer of a place will be made. This gives rise to another problem, that of "grade saturation". More and more students are getting high grades at A level, and it is becoming increasingly difficult to distinguish between the most able students. In my own subject, mathematics, something like 28 per cent of candidates get an A at A level. One way to overcome this would be to split the grade A's at A level by the introduction of an A* grade.

In the past, the use of special papers or Step examinations have been used to try to discriminate between extremely good applicants. However, the best private schools have the extra resources to prepare their students for such examinations whereas poorer state schools do not. The government's current solution is to introduce "world-class tests" to identify the most talented students in every subject. We have yet to see any syllabuses for such tests, and it is difficult to see how these will be any more successful in passing "the private school test".

To be positive, I would like to point to an example from my own discipline. Every year thousands of students enter for the various UK Mathematical Challenges. These tests are genuinely mathematically demanding and fun. I would suggest that a "record of achievement" in tests such as these would be a very powerful extra indicator of mathematical ability over and above A level. It may be possible to replicate this in some other disciplines so that one could track the excellence and development of students over a period of time. In the United States, the initial sifting of students for the Ivy League universities is often done by more generalised national verbal and mathematical tests called Sats. I support the call of my own university for something similar to be tried in this country to give an extra indicator of potential and all-round intelligence.

Now back to the A-level results.

Bill Stephenson is senior lecturer and admissions tutor in the department of mathematics, University College London. He writes in a personal capacity.