Education CommitteeFurther written evidence submitted by AQA (Annex C)

The first point relates to whether we hold data regarding those candidates in receipt of free school meals. The short answer is that we do not. This information is, however, held in the Department for Education National Pupil Database. It would therefore be possible for awarding body data to be matched to this data to allow investigations to be conducted. We have not conducted such analyses at AQA.

The second point arises from my concern that there is the potential for misunderstanding regarding the sharing of data between awarding bodies. Results data are shared between the awarding body technical teams and Responsible Officers solely to support the maintenance of standards over time and between boards. The key instances when data are shared are:

during August, following the summer examination series but prior to the issue of results, when provisional results data are also shared with Ofqual, but not made more widely or publicly available. This allows the provisional results at qualification level (compared to statistical predictions based on the prior attainment of the entry) to be challenged by Ofqual and by fellow awarding bodies. This can very occasionally expose differences in the application of the standard which can be corrected before results are issued. Sharing between boards aids transparency and understanding. These data have not been published but would be available on request from Ofqual and were the subject of a previous note to the committee written by Michelle Meadows and sent on 26 January 2012;

following the summer examination series but after results have been issued. This allows a post hoc analysis of inter-awarding body standards using a different methodology to that described above. As such it helps identify any instances of residual misalignment. Where differences are exposed these can be corrected before the next examination series. The detail of this analysis was also covered by the note to the committee written by Michelle Meadows;

for the purpose of inter-awarding body comparability exercises. These often have a strong qualitative element involving experts’ judgements of the quality of candidate performance. These qualitative data are strengthened by analysis of quantitative data. For example, complex statistical modelling can be conducted to investigate whether differences in standards exist once school effects are taken into account; and

to allow the evaluation of new technical or operational techniques, for example when a new grading structure is introduced (eg A* at A level). To conduct such investigations at a national, inter-awarding body level is more powerful than to use an individual board’s data. The technical expertise to conduct such investigations lies within the boards, which makes sharing necessary.

The sharing of data follows a strict JCQ protocol (a copy of which is attached here) which prevents their being used for purposes other than those intended. In particular, this explicitly rules out the use of data to gain an insight into an awarding body’s entry pattern, and the sharing of data with any commercial or marketing function of the board is not allowed. In AQA, strict divisions exist between the technical, standards and marketing functions.

A final point to make is that Ofqual can and does request any of the above data from the awarding bodies.

April 2012