Conclusions and full recommendations

There is a host of information and many perspectives to be gleaned from looking at how computing education is handled in the latter stages of secondary education in the four nations of the UK.

There are different definitions and approaches, for example Northern Ireland aiming for a digital technology space, recognising the technical and creative tracks this may lead to. Others seem to be tackling the issue from the perspective of a tighter definition of Computer Science / Computing Science, with the curriculum and qualifications in England looking more tightly linked to this approach than elsewhere.

Participation rates appear to be holding up or growing, although it is harder to comment in some cases - such as Wales and Northern Ireland - where new curricula and qualifications are bedding in. In England, Computer Science entries have grown to replace the older ICT qualification but may have reached a plateau and are still below the 2014/15 peak of 98,908.

The Computer Science GCSE is available in around 77% of secondary schools in England, but this figure can mislead as, we know that the percentage of pupils that the qualification is available to is actually far higher. Scotland could until 2021 be seen as an outlier here, with entries for Computer Science falling at all levels. However, this may, to some degree, be being offset by uptake of a wider range of newer and digitally-focused awards and an upturn in mainstream entries.

This is important, as many young people have to make choices that do not involve typical A level or Higher routes (for example) and their potential presence in a digitally-skilled workforce (or in apprenticeships, or further and higher education) could be notable. However, obtaining a clear, longitudinal and comparable set of VTQ perspectives is challenging. What we can see is that there appear to be very few consistent patterns or trends in the data relating to uptake and awards across to four nations:

• England - declining popularity of ICT / digital VTQs at RVQ Levels 2 and 3. Some interest in coding-based options and business-related approaches. The ratio of male:female participation / awards worsens markedly from level 2 to Level 3.
• Northern Ireland - no areas of sustained increase. No perspective on male:female participation.
• Scotland - some interest growing in computer games development and business-related ICT awards, and a narrowing of the gulf between male and female uptake of these options (although this is not the case for the increasingly popular cyber security).
• Wales - generally declining participation in ICT / digital VTQs, possibly underpinned by the move away from the ECDL options. ICT for practitioners at RQF Level 3 is growing a little, but there is still a poor balance between male and female participation.

The scale of uptake of VTQs across the four nations is, by and large, small (and reducing) by comparison with the numbers taking more academic options and many of the large number of awards available are arguably quite niche.

All of the UK nations in this study have a long-standing problem with the balance of male:female participation in whichever variant of the subject they are teaching and whether in academic or vocational domains. Whilst male:female ratios of 2:1 were not untypical of the older ICT curricula, the move to a more computing-focused (even digital technology-led) approach has seen the imbalance grow: in excess of 10:1 in some cases, but more regularly around the 5-6:1 level.

In subjects such as physics and economics, there is a greater emphasis on role-modelling to try to address long-established participation imbalances, and many of the informal Computer Science endeavours are working very hard to make an impression in this space.^[12] There may be lessons to be learned from some of the VTQs where a small number of topics show a better balance or, indeed, from the Computer Science developments in Ireland. This may be linked both to the way the subjects are still perceived and even their status alongside other STEM subjects.

The recent World Skills UK report on the digital skills gap^[82] also resonated with what arose from the VTQ analysis presented here: declining uptake^[13] and a strong gender imbalance. This report also intimated concerns with a mis-match between what the system might offer and what employers and learners want and need. In the analysis presented here, this may be seen as a relatively strong uptake of digital skills related to business and applications, with less interest in creating digital systems.

It is, of course, difficult to compare 2021 and 2020, and with the previous years, as in each case the way in which assessment outcomes have been derived has been different. However, as GCSE at 7 or Grade C (including National 5) and above is often seen as the gateway to study at the next level (A level, AS level, Higher), it will be interesting to see if (as more students in 2021 have these grades) more students and their teachers feel able to tackle Computer Science and its equivalents beyond age 16.

A further issue, which is exemplified by the data from Scotland but is common to all jurisdictions (although the picture in Northern Ireland is less clear), is the relative absence of qualified teachers in schools and the difficulties in recruiting enough new teachers. In England, in spite of the presence of generous bursaries, only 70% of the target from teacher recruitment has been met across the past five recruitment cycles, and the recent uplift may be a one-off response to the pandemic and economic uncertainty.

If, as in other STEM subjects such as mathematics, the shortages are concentrated on schools serving disadvantaged communities, then the issue of equity of access may compound the lack of female participation in the subject.

The Scottish technology ecosystem review^[3] has noted the curriculum on offer in Scotland could be seen as being designed to be delivered by non-specialist teachers, something that may limit the appeal of the subject and send an off-putting signal to students. It is hard to see how the subject can prosper in the absence of a sufficiency of appropriately qualified teachers.

These issues matter regardless of the scale of Computer Science education in the UK, but that statement alone raises two key, and as yet not easy to answer, questions:

• How much is enough?
• Is the drive for digital skills about specialists (for example, coders) or more generally applicable digital skills?

The desire to grow the subject in schools, to accord it status with the other sciences, to make it attractive as a way into careers that might appeal to a good cross-section of young people is self-evident in any of the country-level and UK-wide policy documents.

Indeed, in recent years BCS has been promoting the idea that there is a need to triple the number of young people taking up the Computing qualifications at Key Stage 4 in England (with concomitant changes to teacher supply and professional development)^[14]. Growth of this scale may well be necessary, but as this report notes, there are some common issues with male:female participation that will need to be addressed whether or not numbers develop from their current levels at GCSE, National 5 and beyond.

There may be several issues that limit growth (the relative absence of female students being a major limiting factor), for example the attractiveness of the subject (and an understanding of where it can lead to) - its relevance to ‘people like me’. It is also possible that the route between subject levels is seen as more forbidding or challenging than it should be.

The participation (by cohort) in England has remained quite static at around 13% over the past five years - broadly typical of the other nations, with the exception of Wales where the proportion of the cohort taking ICT remains almost three times than that taking Computer Science (7%). Equally, the ratio of GCSE:A level or National 5:Higher entries is quite revealing in terms of ‘conversion’ of an initial qualification into a higher award. This has averaged about 7:1 in England in recent years whilst in Scotland it is about 2:1 over the same five year period.

There are some indications that young people, armed with good information about the potential for choosing some subjects over others, change their perceptions of subjects and amend their choices (83) so even at the current GCSE / National 5 engagement levels, there may be more to be done (especially in England) to show the career and study options that open up to those taking a higher level computing qualification.

Interestingly, the latest public data concerning uptake of Computing in UK higher education institutions (2019/20) shows that it is the 6th most popular degree group (and second only to Engineering in terms of the suite of STEM subjects) with 34,945 first year undergraduates. This is considerably more than the 22,699 students who emerged from UK schools that year with higher-education-relevant computing qualifications. However, as in the school sector, the male:female student ratios have been stable at around 4:1 since 2014/15^[84].

Overall, whilst there are still questions about what sufficiency looks like, and how to make the supply of students grow and diversify within the school sector, there are also ongoing challenges in addressing the balance between school, and further and higher education supply (let alone apprenticeships). Addressing these issues of attractiveness, scale and aim, particularly with regard to the evolving labour market, are questions BCS and a range of partners and stakeholders may wish to return to.

Gaining a useful perspective on the size and shape of the academic, vocational and technical qualifications market is challenging. For example, posing what are superficially straightforward questions about the number of students applying to take any of the vocational awards, and understanding how representative the students are, has unearthed a number of difficulties that will need to be addressed for any future analysis.

The VTQ datasets, in England, Northern Ireland and Wales cover potentially different age groups of students and rarely offer any data more granular than the total number of students taking each award in any year. This is less problematical with access to academic outcomes data but examining issues beyond male:female participation (for example, access across regions) is still challenging with publicly available data. This makes understanding equity of access, diversity and inclusion issues problematical.