Education CommitteeFurther written evidence submitted by AQA

Briefing Note:

Why you cannot apply the comparable outcomes approach in the early examination unit awards

The key reasons why it is not possible to apply the statistical prediction used to guide the summer qualification-level awards to the earlier unit awards are:

1. The candidates entered for early units are less mature, both developmentally and academically, than we would expect them to be by the summer.

2. The candidates entered for early units may or may not be as well prepared as those entered in the summer.

3. The candidates entered are an unrepresentative sample of the final summer entry, and we cannot assume that they ought to perform so as to achieve the same value-added relationship from KS2 as all candidates do. We do not have sufficient information to adjust the predictions to take this into account.

4. If we apply the qualification-level statistical prediction to each individual unit, when we aggregate all the units together we would not meet the qualification prediction. The outcome would be severe at the higher grades and too generous at the lower grades.

The way in which units aggregate together is influenced by ‘regression effects’. The more units there are making up a qualification, the greater these effects. The weaker the correlation between candidates’ performance on the units is, the greater these effects. These effects cannot be known in the early unit awards.

September 2012