Some 2,700 pupils took part in a five-year study of the SAT exam, which is used in the US to help decide admission to colleges.
Researchers from the National Foundation for Educational Research (NFER) examined the students’ SAT scores and how these related to their GCSE, A-level and degree results. They found that A levels were the best predictor of degree class.
There was no evidence that SAT scores could help identify students “with the potential to benefit from higher education whose ability was not adequately reflected in their prior attainment”, says the final report from the study, which was funded by the Department for Business, Innovation and Skills, the NFER, the Sutton Trust and the College Board.
Nor did the SAT results help distinguish between applicants who achieved straight As at A level.
The authors of the report, Use of an Aptitude Test in University Entrance: A Validity Study, say the results suggest that the use of admissions tests by universities should be investigated.
They also say that variations in the reliability of students’ predicted grades at A level might mean that a post-qualification university application system would be “more equitable”.
In addition, the study found that students from comprehensive schools were likely to gain higher class degrees at university than students from independent or grammar schools with similar A levels, a finding that backs up previous studies.
Students attending highly selective universities were also less likely to achieve a high class of degree than students at less selective universities with similarly good grades.
The findings suggest that “it is more difficult” to obtain a first-class honours degree in a highly selective university, the study says.
Sir Peter Lampl, chairman of the Sutton Trust, said he was “disappointed” that the SAT did not “provide an extra tool in helping to identify talent among students from less privileged homes”.
However, he added, the findings “provide further evidence that universities are right to take into account the educational context of their students when deciding whom to admit, alongside other information on their achievements and potential”.
A previous, smaller, pilot study in the UK had found that SAT scores were only modestly associated with A-level grades, suggesting that the SAT test warranted further study.
In the US, there is no national curriculum and no nationally recognised academic qualifications equivalent to GCSE or A level, so university entrance is based on exams such as the SAT and a student’s high-school grade-point average.