The first few years of any new qualification are carefully monitored, so your GCSE students haven’t been hurt, says Lesley Davies
My overriding priority as a college principal, when I think about my students’ exam results, is that everyone is treated fairly. This year we were all grappling with major changes to most qualifications, particularly GCSEs. How could we as college leaders be sure that our 2017 GCSE students weren’t disadvantaged as guinea pigs on the new 9-1 GCSE qualifications? And were our resit students doubly disadvantaged because they sat a different A*-G qualification 12 months before?
Before moving to Trafford, I was the responsible officer for Pearson’s Edexcel qualifications, so I know how the exams industry deals with these issues. Grade boundaries for GCSE, AS and A-level qualifications are set using an approach called “comparable outcomes”. This means that, at national level, if one year’s students are of a similar ability to the previous cohort, we would expect similar results in a particular subject.
The experts who write exam papers always try to make sure they are at the same level as previous years, but regardless of whether exams turn out to be slightly harder or easier from one year to the next, the system aims to ensure that students receive the grade they deserve, whichever year they sit their exams. There have been well-publicised plans to align key lettered grades to the new numbers for reformed GCSEs, so an A is aligned to the new grade 7, a C is aligned to the new grade 4 and G is aligned to the new grade 1.
At the start there can often be a dip in cohort performance
Comparable outcomes is not a statistical fix, whereby five per cent of students get a 9, and 10 per cent an 8 and so on; exam boards use statistical information about the students sitting their exams and historical data about results in each subject area, and balance it with the views of subject experts to set grade boundaries.
During this period of significant reform in GCSE and A-level qualifications, comparable outcomes is ensures there is stability in the system and creates an anchor between the old qualifications and the new.
As a reformed qualification beds in over time, factors such as familiarity, growing availability of professional development and support resources, increased teacher confidence, greater numbers of past papers and so on, can all contribute to improvements in cohort performance. But at the start, there can often be a dip in cohort performance, known as the “sawtooth effect”.
Ofqual has recently published a study into patterns of performance seen after the 2010 and 2011 reforms to GCSEs and A-levels, which showed that changes in average grade boundaries roughly follow the expected sawtooth pattern. Students and teachers took around three years on average to become familiar with the content and style of the new tests.
In the summer of 2017, we saw the first wave of reformed GCSE Maths and English and reformed A-level qualifications assessed for the first time, and these provide a useful example. Grade boundaries were in some cases set slightly lower than in the previous year. In the case of GCSE English and Maths, we had nearly 1,000 students sitting these exams, and significant changes to level of demand, content and assessments impacted on performance. In the case of some A-levels, coursework no longer contributed to the final grade, while in subjects such as science and psychology, more mathematical content was included.
Without using the principle that at national level roughly the same proportion of candidates should achieve each grade as in the previous year, a significant number of learners would have suffered from that lack of familiarity in the system – and might perhaps even have lost their university place as a result.
So we might see grade boundaries in the first couple of years of a new qualification set lower than in subsequent years, but this should build your confidence that the system will ensure students are not disadvantaged.
Lesley Davies is the principal of Trafford College