Does it matter what we study in school?
A provocation paper
Professor Peter Mandler FBA, Professor of Modern Cultural History, Gonville and Caius College, University of Cambridge
‘Even a decade ago, young people were told that maths and the sciences were simply the subjects you took if you wanted to go into a mathematical or scientific career, if you wanted to be a doctor, or a pharmacist, or an engineer. But if you wanted to do something different, or even if you didn’t know what you wanted to do, and let’s be honest — it takes a pretty confident 16-year-old to have their whole life mapped out ahead of them — then the arts and humanities were what you chose. Because they were useful for all kinds of jobs. Of course now we know that couldn’t be further from the truth, that the subjects that keep young people’s options open and unlock doors to all sorts of careers are the STEM subjects: science, technology, engineering and maths.’ (from a speech by Nicky Morgan, 2014)[i]
Politicians have always been pretty free in offering advice to adolescents about what they should study, based more on their own idea of ‘national needs’ than on what might be best for the adolescents in question. This extract from a speech by Nicky Morgan, then Education Secretary, in November 2014 is in that respect typical, though it gives a very misleading picture of what ‘young people were told’ over the last fifty years. In fact they have consistently been told by the likes of Nicky Morgan that they should be studying the STEM subjects because that was what was best not only for the individual but also for the nation’s economy. And yet over nearly the whole of the same period fewer of them — as a share of the total — have been choosing STEM subjects when they have had a choice, that is, in the English system from about age 13 onwards (when they chose O-level and now GCSE subjects) and in the Scottish system somewhat later. This trend, common to many developed countries yet hardly noticed, has been a natural consequence of widening participation in post-compulsory education. It suggests that young people have different ideas about what is in their own interest than politicians do, and have followed those interests without obvious harm to national interests. Putting young people’s choices at the centre of our understanding of education gives a very different view of the causes and consequences of education than a constant harping on politicians’ (and lobbyists’) rhetoric.
In the 1950s, only a very small proportion of young people took public examinations — O-levels and A-levels — the majority in science subjects. Most boys were taking only science at A-levels and most girls only arts; the boys were then more likely to go on to university, where they continued to study science (science degrees accounted for nearly 60% of the total around 1960), and girls more likely to go into teacher training.[ii] Around 1960, however, the balance began to shift. With the rising desire and ability to participate in post-compulsory education, more and more boys and girls presented themselves for O-levels and A-levels. These new entrants were more likely to choose arts than science subjects. By the mid-1960s this ‘swing away from science’ was well-recognized as an international phenomenon. It continued even after the enthusiasm of the 1960s for more and more education wore off in the 1970s.[iii]
Now new subjects were on offer that weren’t obviously categorizable as ‘arts’ or ‘sciences’, a miscellaneous collection of ‘social studies’ that included social sciences (Sociology, Psychology) and apparently more ‘vocational’ subjects (Law, Business, Communications). In addition, creative arts subjects that had not formerly been examined in school — Art, Music, Drama — were added to the curriculum and gradually introduced into higher education as well. In contrast, few new science subjects emerged. In higher education, the only new science subjects were those ‘subjects allied to medicine’ such as Nursing that helped to compensate for the declining share of medical degrees. But overall, accelerated by a resurgence of participation from the late 1980s, the share of science continued to decline, to below 40% of degrees by the early 2000s. The absolute number of A-levels awarded in Physics hasn’t grown in fifty years, while the number of A-level candidates in all subjects has trebled.
The ‘swing away from science’ has been a consistent concern amongst politicians and industry (especially engineering) lobbyists since it first manifested in the mid-1960s. A national commission was appointed in 1964 to diagnose and correct it. The Dainton Report which resulted in 1968 agreed that the swing threatened a shortage of ‘qualified manpower’.[iv] It blamed the peculiar pattern of early specialization characteristic of England that made people choose between arts and sciences at too early an age. It didn’t recommend state intervention to halt the swing, since in a liberal and increasingly democratic society such as 1960s Britain ‘the individual’s choice and the factors which bear upon it have become critically important’. But it did recommend giving as much advice and information as possible to young people to persuade them to stick with STEM, inaugurating a 50-year tradition of steering to STEM, precisely opposite to the trend Nicky Morgan claimed to have detected.
In fact, as critics pointed out at the time, early specialization had little to do with the swing away from science. Scotland with its later specialization was experiencing exactly the same phenomenon as England and Wales.[v] Furthermore, research aimed at combatting the swing revealed some surprising facts. Britain, far from lagging behind its economic rivals in STEM education, already had much the highest proportion of STEM entrants to higher education in Europe, casting doubt on the alleged shortages of ‘qualified manpower’. In any case, the swing was occurring in most developed countries except France.[vi] Furthermore, research began for the first time to focus on why students were swinging away from science — what determined their choices and why they weren’t choosing science.
Perhaps unsurprisingly, survey after survey found that students chose subjects because they liked them and were good at them. Career prospects hovered in the background but were rarely prime determinants. Even at 18, one representative sample of university entrants in 1996 found that only 18% had clear career objectives. 85% of arts and humanities students but also 69% of social science and 61% of Maths students said that their primary motivation was enjoyment of the course.[vii] Behind these preferences for enjoyment and ability, naturally, lay other determinants rooted in students’ background and environment. Boys tend to enjoy and excel at science, girls at English or languages — or, to press the point, boys at Physics, girls at Biology — for reasons that must largely have to do with upbringing. But those background and environmental influences change over time and in this way bigger social and cultural changes have a direct impact on education by changing what children enjoy at school.
The biggest change has been in gender roles. From the late 1960s, girls felt more entitled to choose subjects of any kind and to enter for exams. The number of boys on A-level courses increased 8% in the late 1960s but the number of girls by 30%.[viii] This gender swing continues to the present day, so that whereas only about a quarter of undergraduates were women in the 1960s today they form well over a half. Although their propensity to study science has grown, they still do so less than men so that the gender swing entails a swing away from science. To some extent this has benefited the arts and humanities. But it has also benefited the social studies, especially as the 1960s and ’70s also saw teenagers citing a growing ‘desire to help the community directly through social action’ to explain their choice of subjects.[ix] Psychology, which had been a male-dominated subject when it entered curricula in the 1960s, is today overwhelmingly female-dominated: three-quarters of A-level candidates are women.[x]
Widening participation in other respects has also had dramatic effects. Children whose parents had little education themselves are now much more likely to take A-levels or other post-16 qualifications and to go on to university. Their own early-years attainment levels tend to start lower, and by the time they have to choose subjects they are more likely to shy away from traditional ‘academic’ subjects in both arts and sciences and to gravitate to new subjects like social studies. Psychology is again among the most popular ‘new’ subjects after GCSE.[xi] Some of this has to do with ‘difficulty’, but some of it has to do simply with prior attainment — sciences and languages are ‘cumulative’ subjects in which one does better if one starts earlier.[xii] It is no coincidence that both have been in decline in a period of widening participation when more students enter the subject game relatively late in the day. But widening participation is not the only factor in explaining the swing away from science. As girls’ growing preference for ‘social action’ suggests, other changes in young people’s self-image and sense of purpose also have an effect, even on very early choices. In the 1960s and ’70s the most significant trend was in favour of ‘people’ over ‘things’.[xiii] Since the 1990s, the trend has deepened in favour of ‘self-realisation’ and the development of peer-group identities.[xiv] Psychology has again benefited greatly.
All of this freedom of choice may seem understandable, and yet at the same time rather self-indulgent. Teenagers can’t be expected to be ‘econometricians’, as the sociologist John Goldthorpe has pointed out, calculating as some economists imagine they do the realistic lifetime earnings returns to the choice of specific subjects. But perhaps they should be paying more attention to such things, both in their own economic interest and to serve the needs of the labour market. It was with that in mind that Nicky Morgan was making her pitch in 2014, and since 2014 the drumbeat has deepened, as government generates more data on the earnings returns to studying particular subjects at particular universities through the LEO (Longitudinal Educational Outcomes) database. Besides, as we are told — most recently, in the Augar Review — skills shortages in ‘STEM-related areas and the skilled trades’ are holding back economic growth.[xv] Why wouldn’t we all want a swing back to science?
It’s fair to give the student’s eye view for a change. Are ‘STEM-related jobs’ really their best bet? It’s far from clear that these ‘STEM-related’ shortages actually exist. Augar’s own data shows that there are vastly more vacancies in ‘shortage’ areas such as business, wholesale and retail, and leisure than in specifically ‘STEM-related’ areas such as IT or manufacturing.[xvi] Today’s economy is not more ‘STEM-related’ than it used to be — it’s much less, mostly because engineering and manufacturing are much less significant. If there is a ‘skill bias’ in today’s economy, it’s not towards STEM-specific skills but a much more generic set of managerial, analytic and human-relations skills which can be cultivated by just the wide range of ‘social studies’ (and arts and sciences) subjects that most students choose.[xvii]
What about the wage premium that STEM students do undoubtedly get? Isn’t that a sign of a ‘shortage’ that eager young people ought to rush to fill? There’s certainly evidence that this view has been gaining traction in the last few years since Nicky Morgan’s speech. After the 2008 economic crisis, and especially the 2012 tuition fee hike, for the first time since the 1960s the swing away from science has been halted and even gone into reverse.[xviii]
How long this trend will persist or what will be its effects — or even, really, what are its causes — is hard to say. But if it is motivated purely by the search for a wage premium, irrespective of enjoyment and ability, then its effects are likely to be perverse. There is much evidence that the wage premium for STEM graduates derives not from ‘STEM-related’ employment — after all, only about a quarter to a third of STEM graduates enter ‘STEM-related’ employment.[xix] It is more likely that employers use a STEM degree as a screening device to find the high levels of general skills that they need for the widest variety of jobs. Only about a quarter of jobs need skills specific to a subject.[xx] To that extent, Nicky Morgan was right. STEM subject choices do send employers good signals for a wide range of jobs. But if more people take them up, that screening value will diminish. Other signals — such as a degree from an elite institution — will come to work better. The STEM effect will wear off.
Worse still, there is evidence that if people choose subjects not on the basis of enjoyment and ability then they lose out even in financial returns, not to mention job satisfaction.[xxi] In the longer run, it’s much better to follow the subject that inspires the focus and the hard work that stems from enjoyment and ability and leads to both better job satisfaction and better job performance. It doesn’t matter all that much what subjects you choose. Once you have the basic literacy and numeracy skills that do need sedulous cultivation in earlier years, as a foundation for higher-level achievement, then most academic subjects cultivate the general skills that employers need in the ‘knowledge economy’. Subject choice needn’t and shouldn’t be all about the labour market, even if teenagers could be econometricians and also fortune-tellers of future needs. We could fill all those business, retail, office, leisure, health and social care jobs with reluctant STEM graduates, or we could fill them with graduates who chose more freely and happily from the very wide range of subjects now on offer in post-compulsory education, which serve their full range of passions and interests while also raising the general skill levels in the population. What choices they make will determine the health and welfare not only of our economy but also of our culture and society.
The British Academy has undertaken a programme of work that seeks to re-frame debates around childhood in both the public and policy spaces and break down academic, policy and professional silos in order to explore new conceptualisations of children in policymaking. Find out more about the Childhood Policy Programme.
Subscribe to our newsletter for monthly updates on our Childhood Policy Programme.
[i] Speech by Nicky Morgan, 10 Nov. 2014: https://www.gov.uk/government/speeches/nicky-morgan-speaks-at-launch-of-your-life-campaign.
[ii] Committee on Higher Education, Appendix II (A): Students and Their Education, Cmnd. 2154-II (1962–3), 31; Celia M. Phillips, Changes in Subject Choice at School and University (London, 1969), 10–27.
[iii] Ibid., 37–42.
[iv] Council for Scientific Policy, Enquiry into the Flow of Candidates in Science and Technology into Higher Education, Cmnd. 3541 (1967–8), Preface.
[v] Andrew McPherson, ‘The Dainton Report — A Scottish Dissent’, Universities Quarterly 22 (1967–8), 254–7.
[vi] Andrew McPherson, ‘“Swing from Science” or Retreat from Reason?’, Universities Quarterly 24 (1969–70), 29–43; Phillips, Changes in Subject Choice, 110–11, 115–17.
[vii] Kate Purcell and Jane Pitcher, Great Expectations: The New Diversity of Graduate Skills and Aspirations (Warwick, 1996), 11, 23.
[viii] D. Duckworth and N.J. Entwistle, ‘The Swing from Science: A Perspective from Hindsight’, Educational Research 17 (1974), 49.
[ix] D. Duckworth, The Continuing Swing? Pupils’ Reluctance to Study Science (Windsor, 1978), 32.
[x] Jessica Bowyer and Prerna Carroll, ‘Students’ decision-making about A level Psychology’, Psychology Teaching Review 22:1 (2016), 6–7.
[xi] Kevin Walker, ‘Why do sixth form students choose Psychology? A report of research in one institution’, Psychology Teaching, Summer 2004, 29.
[xii] Duckworth, Continuing Swing, 56–7.
[xiii] Duckworth, Continuing Swing, 13.
[xiv] Ellen Karoline Henriksen, Justin Dillon and Jim Ryder (eds.), Understanding Student Participation and Choice in Science and Technology Education (Dordrecht, 2015), 2, 20–1; Ross Finnie, ‘Does Culture Affect Post-Secondary Education Choices?’, Higher Education Management and Policy 24 (2014), 69–70.
[xv] Independent Panel Report to the Review of Post-18 Education and Funding, CP117 (2019), 25–6.
[xvi] Ibid., 50. Augar reconciles this contradiction by emphasizing the proportion of vacancies within each sector rather than the relative size of the sectors. There are many more vacancies in non-STEM ‘shortage’ areas, with the exception of health and social care.
[xvii] Yujia Liu and David B. Grusky, ‘The Payoff to Skill in the Third Industrial Revolution’, American Journal of Sociology 118 (2013), 1349–51, 1357–9;
[xviii] DES, Education Statistics for the United Kingdom (1967–78); Statistics of Education, vol. 6 (1979); Universities’ Statistical Record, University Statistics, vol. 1 (1980–92); Higher Education Statistics Agency, Higher Education Statistics for the United Kingdom (1993–97); HESA Online: https://www.hesa.ac.uk/data-and-analysis/releases (1997–2017). Great Britain only 1980–92. Data for 1993–2004 are anomalous as the statisticians took a while to reconcile the different degree structures of the pre- and post-1992 universities.
[xix] Emma Smith and Patrick White, ‘Where Do All the STEM Graduates Go? Higher Education, the Labour Market and Career Trajectories in the UK’, Journal of Science Education and Technology 28 (2019), 26–40; Emma Smith and Stephen Gorard, ‘Is There a Shortage of Scientists? A Re-Analysis of Supply for the UK’, British Journal of Educational Studies 59 (2011), 159–77.
[xx] Institute of Student Employers, Annual Student Recruitment Survey 2018 (London, 2018), 31.
[xxi] Zachary Bleemer, ‘Wage-by-Major Statistics: Transparency to What End?’, Inside Higher Ed, 24 Jun. 2019.
