Those AQA English Numbers Explained

What the hell just happened?

The January exam produced lower marks than expected. AQA looked at these results and, with this being a new qualification, felt that in order to make the results from this new GCSE comparable with previous years they needed to alter the grade boundaries. Thus, they set very favourable thresholds. In order to get a C, a student needed to obtain 54% of the marks available in the Foundation paper and just 51% in the Higher paper.

What AQA failed to grasp was that the lower performance in January was not a result of teachers and pupils struggling to meet the demands of the new spec, but rather a result of schools deciding to enter their borderline students in January. This would afford the opportunity of a resit in June. Therefore, the calibre of the students entered in January was considerably lower than when the majority of the year group sat their exam in the summer.

Once the results from June came in, AQA realised they had a problem. If they were to apply boundaries comparable to those in January there would be a huge improvement in English results across the year. This would lead to further accusations of grade inflation and would seriously undermine the validity of their new specification.

How did AQA overcome this problem?

This is the travesty and where accusations of butchery are well founded. In order to overcome the problems they created in the setting of the boundaries in January they needed to adjust the summer grades. Rather than balancing the grade changes across Higher and Foundation they primarily attacked the Foundation paper.

This June you needed 11% more marks than in January to achieve a C in Foundation as they raised the threshold to 66%, but only 4% in Higher where the C threshold was set at 55%.

Even ignoring the January/June discrepancy, there is an 11% difference in the marks needed to get a C in the parallel higher and foundation papers this June.

Are the board allowed to do this?

In the specification the board have published maximum grade boundaries at 60% UMS (Universal Marks Score). This would make it appear that by setting a boundary at above 60% they are in contravention of their published maximum. However, the UMS is the number which is arrived at after the students’ actual marks have been adjusted in accordance with the grade boundaries. Therefore, the UMS will always show 60% as a C as it has already been adjusted.

In short, as far as marks are concerned, I think the board can do what the bloody hell they please.

So does the legal challenge have any hope?

I don’t know law, but I don’t think there is any way to challenge the board on the basis of their right to change boundaries. Rather, as the legal impetus grows, the means of a challenge is on the effect of these changes. Because the main attack on the grades was on foundation students, if the suit can claim that there is a particular profile of foundation student who tends to be from a poorer or more disadvantaged background then these changes will be in contravention of Equal Opportunities legislation.

Why have some schools been affected more than others even if they both sat their exams in June?

Simple. The higher the proportion of foundation entries you have, the more you will have been affected.

How are the final marks calculated?

For each question on the exam pupils are awarded a mark. This is the raw mark which AQA publish as part of their Question by Question breakdown on their Enhanced Results Analysis service on their website.

A complex formula then works out within which of the grade boundaries this falls and how far into the boundary it comes and from the results of this produces a Universal Mark Score (UMS) for each mark.

Therefore, the resulting UMS has already been adjusted in accordance with the grade boundaries.

So using UMS marks in schools has been a waste of time?

Only now. The specification’s published UMS boundaries are predicated on a boundary threshold of 60% for a C 70% for a B etc. 60% has traditionally been a high cut off, therefore using these thresholds has been a good yardstick for safety and could often result in some pleasant surprises as the thresholds were set lower.

This year however, the Unit 1 Foundation has smashed the 60% threshold for the first time I can remember and has shown that the UMS is only a guide.

Is there a simple way of working out the effects different grade boundaries will have on a set of marks?

No. Because the changes in the boundaries are not evenly spread across the grades (for instance, a C may cover 12% of marks where as a B may be 8%), there is no uniform number which you can multiply marks to find out how they would change. I have written a formula for Excel which will do this accurately but it is a side of A4 long and refers to a number of different sheets and workbooks in order to work effectively.

Where do we go from here?

Who knows. Strategic planning in order to do the best for our pupils has been reduced to trying to second guess what the boards are going to do. Do we enter our new Year 11s in January on this last opportunity or will they be overly strict in order to compensate for this year? Do we abandon Foundation as it has been decimated this year, or will this too change to become more fair? Do we change boards?

The remarkable cock up by AQA this year might have the effect the government have always wanted. We can no longer play the percentages and selectively adjust our focus as we have done in the past. Things have been blown apart. We must simply focus on teaching well and hope for the best.

Leave a comment


  1. aj

     /  August 25, 2012

    given all this, it seems bizarre that AQA state “Our Principal Moderators’ judgement was that the individual grade boundaries for the summer needed to either stay the same, or be increased. Our boundaries therefore changed by between 0 and 3 per cent”

    for English grade C the changes look far more than the stated 3%?

  2. My only explanation is that this figure is an average change to the boundaries across all grades in the higher and foundation papers. It is probably true overall but obfuscation at its worst.

  3. Faye

     /  August 26, 2012

    What is more worrying for my pupils (and millions more I suspect) is that even if they got a C on the highly inflated F paper, their controlled assessment and speaking and listening marks (60% of final grade) have also been subject to changing boundaries. For example, a pupil who achieved 43 on unit 3 (written controlled assessment) was in any other series a C. Now 46 is the boundary for a C grade. How can AQA justify changing controlled assessment boundaries? They have a moderation process to change a centre’s marks if they feel marking is too generous. We must call for a reversal in the boundaries for unit 2 and 3 at the very least! Then worry about fighting for the paper boundaries.

    • I agree that this is again completely unfair. However, at least these grade thresholds hover around the 60% figure published in the spec so schools cannot have been completely screwed in their projections. Also because there are no H/F issues in Unit 2&3 at least the board’s attempts to rectify their mistake affected all entrants in June equally. I fear this may make it much more difficult to force the board’s hand as it is harder to make an Equal Opportunities case.

  1. Transcending GCSE English Grade Boundaries - Christopher Waugh

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: