Which courses has your provider stopped offering over the last two years? Which courses are new in that time?
Is there churn in business studies and politics, while sports science pulls away from the field? For a regulator with an interest in the shape of the subject area market, these are questions of interest.
The OfS’ investigations of the state of the market for subject areas spurred me to break out the unistats data – and in doing so I realised there was another important question we could start to answer.
Providers offer and cease to offer courses all the time – and these decisions are a useful way to understand the way that the applicant marketplace is moving. Courses exist to inspire student enrollment – and a course that does not recruit well may not be offered again. But providers may also discontinue provision for wider strategic reasons, making the decision to close departments and leave subject areas for financial reasons.
Likewise, the choice to offer a new course may come from market intelligence – but it could also equally stem from academic interest, local employer needs, or a need to redeploy existing capacity.
Back to terms
Andy Youell’s seminal work – “What is a course?” – was published in December 2011. It is a fascinating read, exploring just how hard even the most fundamental data questions can be to answer, but for our purposes we can start with his definition;
there is a broadly common model of a student registering on a thing that is made up of components (which might include sub-components) and which is defined by some sort of outcome
A “course” could also be defined as an entity described in the unistats dataset as a course. This causes us some problems as a change in course attributes results in a new course which is nearly identical to an existing course. A full time BSc (Hons) Computing is not the same course in unistats as a part time BSc (Hons) Computing, or as a full time BSc (Hons) Computing with a compulsory foundation year, or as a full time BSc (Hons) Computing (Cryptography) – even though students may sit in the same lecture halls taking the same course.
So our unistats definition could be expressed as:
A thing that is offered for prospective students to apply to
This means that any insight that we glean from unistats data tells us more about marketing than academic or strategic decisions. Useful information in itself, but it is important to recognise this limitation. Equally, subject (SBJ) tagging is not universal within the data, there are sizable chunks in both graphs that are not linked to a specific CaH level three subject.
Paths less taken
This visualisation shows the 13,306 courses that were represented on unistats as of January 2019 (so primarily for the 2020 UCAS application cycle) but were not represented in unistats as of October 2020. (It will also include any courses that have had their course ID changed in the data – it is generally accepted good practice to preserve identifiers for courses).
Sheet one shows these courses by provider for each subject area, sheet two shows them by subject area overall, and sheet three lets you look at an individual provider.
Lost courses are split widely between subject areas – business studies, politics, and sociology courses are most likely to be discontinued, though it is important to recognise that many of these are courses that span multiple subjects (for example “politics with sociology”). The Universities of Liverpool and Derby in particular appear to have cut large numbers of joint honours courses.
The individual institutional view does let you see some evidence of strategic movement – some providers are notably moving away either from (high-cost) engineering subjects and from (competitive) creative courses.
This time we’re looking at the 10,977 courses represented on unistats as of October 2020 (so primarily for the 2021 application cycle) that were not in the data as of January 2019. (the point about course ID changes also applies). The three sheets are organised in the same way as previously.
Sports science seems to be a key point of movement – nearly 400 new courses are on offer in that subject area since 2019, though a similar amount have been lost over those two years . On a macro level most increases are in business, creative arts, and social studies – a sign perhaps that pro-STEM messaging is not cutting through to the applicant market. There’s a notable increase in language provision (with UCL as a key contributor here) which perhaps suggests that Brexit is increasingly an influence on what marketing staff believe prospective students may be thinking about.
I’m sure many will be interested in how the subject offer is changing at course level – and that this may include regulators. So here’s a simple look at which subject areas have grown (in terms of the number of courses offered) and which have shrunk. The graph at the top shows change between Jan 2019 and October 2020 at CaH top level, click on a bar to drill down to level 2.
It probably will surprise no-one that a growth in student numbers for subjects allied to medicine (particularly non specific courses and nutrition) is mirrored by a growth in the number of courses – it may be more surprising that historic growth in the creative arts (particularly design studies and music) has also seen a growth in course numbers.
Numbers of courses are falling in the social sciences (specifically politics and sociology), in history and also – surprisingly – in business studies.
This same time period has seen ministers take an active position on “low quality” courses at odds with the way that the Office for Students was set up to grow a market that could make these decisions via a few enhanced indicators like TEF. Despite my best efforts we’ve not yet reached a definition – but it may be interesting to know how these patterns play into perceptions at DfE. This is, of course, far from perfect data (I’m looking at KISCOURSEID, so anyone who just bought a new student record system will be giving me false positives) but it is the best we have right now.