New rules on exam readers could become a headache for schools

7 Aug 2019, 13:12

A change in the rules on reading assistance in exams for pupils with special needs could cause a jump in the number receiving such assistance next year. Can we quantify such a potential rise and what will it mean for schools, asks Barney Angliss.

JCQ’s decision to change the rules on what it requires from schools who want to use readers for their pupils in exams is more in tune with the Equality Act’s requirement for “reasonable adjustments” but it has immediate practical and resource implications.

This could grow the percentage of pupils receiving assistance from 7.3 per cent last year to 15.9 per cent

The proportion of exam candidates requiring the help of a reader (or reading software) was 6.8 per cent in 2016-17 – almost exactly what would be predicted in any cohort, using the regulated criterion for such help at the time (a standardised reading score below 85, assessed by a qualified specialist).

However, the proportion being awarded reading assistance last year rose above predicted, to 7.3 per cent. The percentage for 2018-19 won’t be published for some months yet; but it will be the last cohort for whom that cut-off applied.

From September, candidates will be allowed reading assistance simply on the basis of a written note from the centre’s SENCo that they have “an impairment which has a substantial and long term adverse effect, giving rise to persistent and significant reading difficulties” and that the use of such assistance is the candidate’s “normal way of working”.

It can be argued that a simple threshold such as a standard score below 85 isn’t a sufficiently reliable indicator of reading impairment with a “substantial and long-term adverse effect”. It only represents a point 1.5 standard deviations below the population’s mean reading ability (-1.5 s.d.).

It’s a statistical convention rather than a functional measure of disability, but it has been an accepted benchmark in our special needs system for decades and further rises in the proportion of pupils receiving either human or computer reading assistance may be expected now that the rule has been relaxed.

Schools already work very hard to get approval for this type of support, paying significant fees for specialist assessments, allocating a substantial amount of time from their SENCo and examinations officer to get it right (with no-notice inspections of the process during the exam season) and funding the support during lessons and mock exams too.

These “barriers” are often explained to parents as reasons why a pupil can’t be offered reading assistance – and they rest on the “standard score below 85” rule.

Centres will now be under pressure to implement reading support for candidates just outside the current range.

Let’s suppose, for instance, that schools include pupils who have a reading score just a little above the present boundary – say, 89 or below (-1 s.d.), rather than 84 or below – but who, with greater access to the exam paper, could achieve significantly better marks.

This could grow the percentage of pupils receiving assistance from 7.3 per cent last year to 15.9 per cent, more than doubling the number who need to be using a human reader or reading technology as their “normal way of working” in school to justify these facilities and who must be appropriately accommodated for exams (seated away from others if they have a human reader, or at a computer if they use software and headphones).

The increase in support, evidence-gathering from teachers and information to parents would all impact on workload and could cancel out any savings made by reducing the number of specialist assessments previously required to identify reading impairment.

While the change will be good news for some pupils who struggle, what will it do to results nationally? If there is any significant improvement in these candidates’ attainment as a result of being offered reading assistance they were previously denied, isn’t it likely that grade boundaries will be nudged upwards to maintain the statistical validity which regulators demand, pushing the same candidates back down to where they were?

“Why not?” the authorities will say, “it’s our normal way of working.”

Your thoughts

Leave a Reply

Your email address will not be published.