The Telegraph reports on efforts to make A-levels “harder”. My concern is that hard-working students and teachers too often have their work and achievements undermined by speculation about the quality of the system.

Is the problem more fundamental?

According to the QCA, grading swapped from “norm referencing” – where, say, the top 10% get an A – to “criteria referencing” – where you get an A if you meet the criteria. On first inspection, there seem to be two problems with criteria referencing: the criteria and the usefulness of the grade.

It appears that criteria referencing has been chosen so that everyone who meets the criteria gets the grade, and so that year on year, greater numbers of passes at higher grades are a meaningful indicator of improving standards. But surely when, inevitably, you change the criteria, the foundation of each grade is undermined? Surely the question “Are A-levels getting easier?” is only meaningful under criteria referencing? Provided questions range adequately in their difficulty, wouldn’t norm-referencing solve this problem?

Does criteria referencing answer the question universities and employers are asking: how did people fare relative to one another? Competition is a fact and hard-pressed companies can’t afford to make a bad hiring decision, so if A-levels aren’t grading people helpfully, the burden must fall on some other part of the recruiting process, shifting the burden of cost and risk further to the employer.

It seems that criteria-referenced A-levels are having their criteria changed to make them harder, to better grade students, but the changes themselves undermine the system. Does criteria referencing actually represent “progress” and is it time to return to norm-referenced A-levels and harder exam questions at the extreme?

Labour appear to be in yet another pickle, but in the meantime, a heartfelt congratulations to all those who have demonstrated their abilities through this year’s exam results.

read more | digg story

One Comment

  1. Hmm… I see your point, but the problem with reliance on grade norming is that it doesn’t fairly compare ability levels between two people who happened to be in different years at school.

    Both methods are measurement techniques, attempting to measure ‘how good is this person at subject X’?

    Young people appear, on average to have been getting worse over the years, because many subjects are being taught in a watered-down softball way, with over reliance on coursework, and progressively less and less complex and rigorous questions in the exams. All built on an increasingly shaky foundation of basic English and mathematics.

    If that is allowed to continue, then worrying about whether grades are based on comparison within a peer group, or whether based on a constant absolute scale won’t fix anything.