View all news

Reform of AS-levels was based on ‘shaky’ evidence

4 August 2014

Reform of AS-levels was based on ‘shaky’ evidence

The Government’s justification for its decision to reform AS-levels was based on flawed data analysis, according to University of Bristol academics who have shown that one in five students could have their chances to fulfil their potential on a degree course damaged.

Changes to the examination system for post-16 students at English schools and colleges mean AS-levels will no longer count towards full A-level results, altering the basis on which universities can make their conditional offers to applicants. 

Minister of State for Schools David Laws commissioned research to compare whether GCSE results were as good as AS-level results as a predictor of whether students would go on to achieve a 2.1 or above at university. The results, published in May 2013, found that degree performance can be predicted to a similar level of accuracy based on GSCE grades alone.

The study, of 88,022 students who graduated in 2011, was used to justify the Department for Education’s reform of AS-levels, which will act as standalone courses from 2015.

However, when researchers from the University of Bristol took a closer look at the exact same set of data used by the Government – which took a year to be sent under a Freedom of Information request - they identified missing data, sample bias and poor research design.

Professor Ron Johnston, Professor of Geography at the University of Bristol, said: “The research undertaken by the Department for Education to provide evidence as a foundation for the abolition of AS-levels as the first stage of the A-level exam was, to say the least, unimpressive.

“The students included in the analyses were not representative of all who graduated in 2011 – many of those doing science and engineering were excluded as were all of those doing medicine and the great majority of those who attended Scottish schools and universities. The analyses included students who should have been excluded because of incomplete data, the modelling strategy was inappropriate and some of the findings nonsensical.” Read more...


Edit this page