The University of Roehampton has released its latest Roehampton Annual Computing Education Report (TRACER) which provides an analysis of the provision of GCSE and A-level computing qualifications in schools in England, considering factors such as student gender, ethnicity and geographic spread. This is the second report of its kind produced by the University.
- Increasing numbers of schools are offering Computer Science (CS) at GCSE (52.5%) and A level (36.2%), and so now there’s a good chance that a student will find CS on offer at their school (76.3% at GCSE).
- Relatively few students choose to take the subject: at GCSE, only 11.9%, and at A level, just 2.7%.
- Provision though remains ‘patchy’: grammar schools are more likely than comprehensives to offer CS, independent schools rather less so.
- Some local authorities and multi-academy trusts lead the way, and others lagging behind.
- Numbers taking the subject continue to rise, although not as rapidly as in the past.
- At GCSE, the typical CS student is academically strong, mathematically able, likely to be taking triple science (despite CS counting as a science for the EBacc), from a relatively affluent family, and overwhelming likely to be male (even if the smaller number of girls taking the subject do better in the exam). Some schools and local authorities are doing well in addressing the gender gap in CS, but there are 382 ‘mixed’ schools where the CS students are all boys.
- A-level CS remains a niche subject: students typically have good maths grades, but their overall academic performance is not strong. CS is often taken in combination with maths and physics.. Again, students are likely to come from relatively affluent backgrounds, but rather more of these students will be on the school’s SEN register than for most subjects.
- Girls continue to be heavily underrepresented in the study of computer science. In 25 local authorities, all the CS entries come from boys. At GCSE 20% of entries are from female students (1 in 5) and only 10% (1 in 10) at A-level, even though girls do better than boys at GCSE. Of note is that girls no longer outperform boys at the top A-level CS grades, A* and A. Only 34.2% of all females are taking a computing KS4 qualification, compared to 51.2% of all males.
- CS and ICT are quite different qualifications, and thus are taken by quite different students: the latter are (on average) from less affluent backgrounds, weaker academically, closer to a typical mix for ethnicity, and more likely to be female: the decision to remove ICT as qualifications at GCSE and A level, seems likely to result in fewer, and rather less diverse, students overall taking qualifications in computing.
- GCSE and A level CS are hard! At GCSE, students typically get half a grade lower in CS than in their other subjects; at A level, CS grades are also a little lower (about a sixth of a grade) than those students get for their other subjects.
Overall, these findings suggest that we now must rise to the challenge of encouraging (or perhaps allowing) more students to have a go at CS, learning from the good practice in schools, local authorities and trusts that are already succeeding here.
This issue is particularly acute at A level, where less than 15% of colleges or sixth forms have cohorts that the DfE would regard as viable.
Schools entering students for GCSE computer science
There is substantial variation in GCSE Computer Science provision between English regions. In particular London (49.2%) and the West Midlands (47.3%) see the lowest percentage of providers offering the subject, compared to 55.3% of providers in North West. Uptake ranges from 13% in the North West to 10.2% in Yorkshire and the Humber.
Students, schools and female students for GCSE computer science, by local authority
There are some examples of GCSE CS provision and uptake across the country with Bournemouth, Knowsley, Trafford and Hartlepool local authorities (LAs) having more than one in five of their students studying the subject.
However, schools in many local authorities still have a lot of work to do, with low uptake amongst several local authorities, e.g. Kingston upon Hull (3.9%) and Luton (5.1%).
You can read the full report here.