What are the respective impacts of socio-economic disadvantage and prior achievement on student success post-16?

We now have measures which help us to understand and compare both. These data are very useful at institutional level to establish the difference between the profile of a particular sixth form and the national average, both in terms of disadvantage and prior achievement and therefore to contextualise the performance of students in that sixth form.

But what do the national data tell us about the relationship between these factors and student performance overall?

When sixth form colleges are ranked in terms of the socio-economic deprivation of their students and their success rates are divided into quartiles (25% of colleges) and compared, what emerges is…actually not very much.  At AS and A level, only in the least disadvantaged 25% of colleges do students do better on average with a success rate of 88% while students in each of the three other quartiles, including the most disadvantaged 25%, achieve on average a steady 85% success rate. For advanced vocational courses, the relationship is inverted with students in the most disadvantaged quartile achieving 83% success rates on average while each of the others achieve 82%. Neither of these relationships provides any correlation and certainly not in a consistent direction.

The figures at level 2 and level 1 are even more difficult to interpret. From most disadvantaged to least disadvantaged, they are:

Level 2: 87%, 84%, 81%, 88%

Level 1: 88%, 89%, 83%, 86%

Hardly what one would call a pattern in either case.

These data are from the 2012 Ofsted Socio-economic performance indicator (SEPI) report.

So, there would seem to be little evidence for our intuitive sense that more disadvantaged students face more barriers to success and would therefore be expected to achieve lower success rates. While this may be true at 16, it doesn’t seem to be true at 18.

But when we look at the effect of prior achievement, the story is very different. All the available measures of value added confirm a strong link between prior achievement and A-level success rates and grades. In fact, the whole value-added methodology relies on this correlation. Nick Allen’s ‘Six dimensions’ report for example has average AS level success rates ranging smoothly from around 96% for students with an average GCSE point score of 7.0 or more down to below 60% for those students who started with GCSE point scores of 4.7 or below. At A-level, where there has been more ‘weeding out’ of lower achieving students, the correlation is still strong, ranging smoothly from around 98% to around 91%.

Students’ chances of getting high grades at A-level are dramatically affected by their prior achievement. If your average GCSE point score was 7.0 or above, you have an 80%+ chance of achieving a grade A*-B at A level, while at the 4.7 end of the range that chance is barely 15%.

These data are form the Nick Allen ‘six dimensions of performance report’ for 2014 and are overall averages. There are significant variations by subject.

The message seems clear: students who do well at advanced level are those who have already done well at GCSE level and it seems that this deterministic message rolls through the whole of a young person’s schooling meaning that there may well be a strong correlation between the performance of young children at primary school and their eventual sixth form performance. The depressing conclusion is that we’re not really breaking the cycle of disadvantage.

But what about those socio-economic data which seem to suggest no real correlation between disadvantage and achievement at 18?  Well, these should be taken with a heavy dose of contextual salt. The students progressing onto, and entering, qualifications at each level post-16 are only the subset of the whole cohort who have met the entry requirements for those qualifications. They have already crossed an important threshold and are therefore already selected for success. Any impact of socio-economic disadvantage on their success will already have happened and, if it is negative, this will prevent them from even getting to the starting post of the higher qualifications.

If we really want to understand the impact of socio-economic disadvantage on achievement at 18, we should compare the profile of those who do well with the profile of the whole age cohort. When this is done, I suspect a rather more depressing message emerges. There is a lot more that education and society as a whole have to do before we can say that opportunities are genuinely equal for all young people.