A lot of people enter higher education but do not complete their course – and we never see them in sector data.

Put starkly, this looks a lot like failure – either a personal failing or something that has gone wrong with their course or provider. But the simplistic idea of a student “giving up” or a provider “giving up” on a student (or indeed, a student decision that uni wasn’t for them) masks a great deal of possible events and outcomes.

Away from the published figures, having a prompt and accurate understanding concerning which students are no longer studying on the courses they sign up to is important every year. In 2020-21, with rumour of large numbers of students leaving the course or never turning up the importance of having an accurate student roll is even greater.

At the top level, “non-continuation” is a UK performance indicator, one of four that is published annually in benchmarked form. The headline figures describe one particular event – a student who remains in HE for at least 50 days after a “start date”, and does not continue in higher education beyond their first year.

The UK performance indicators saw another lease of life as TEF metrics. With the NSS (which provided three of the six metrics) now out of favour, non-continuation has become one of DfEs favourite metrics (graduate job skill level is the other, something which I’ll get to in the next article in this little series).

What is non-continuation?

When a student “leaves” higher education they tend not to make a sudden decision and then immediately notify everyone that needs to know. The process is best understood as a gradual disengagement. A student may stop attending lectures first of all, or find themselves unable to continue to support themselves financially, or find themselves with other responsibilities and other constraints that stop them being able to participate. Self-definition as a student can be powerful – and it is possible for someone to think of themselves as a student on a course even after they have ceased any interaction with their university.

Universities offer support and advice that is designed to help students before they decide they are leaving – students who have engaged with this and still decided university or their course isn’t for them will be directed to complete a form (like this one for Liverpool, or this process at DMU). At that point, a provider can be sure a student has left – and a reason for leaving can be returned, as required, to HESA (I’ve never seen this published).

Otherwise, a provider will use metrics-based participation data (a register, automated check in, engagement on the virtual learning environment), or even reports from fellow students, to know whether a student is likely to have decided not to continue. Often individual lecturers will pass on likely attrition to a centralised support service. In these cases the university will make an effort to contact the student, offering advice and support. A student determined to leave will be advised to contact the Students Loan Company to ensure that further payments (and thus further additions to their loan balance) do not accrue. Providers are also required to contact SLC in this instance.

Timing is everything

If you are anything like me, that “50 days” number in the UKPI definition made you pause for a second. If term starts, say, on 5 October the official statistics ignore students who leave the course before 24 November. Or – more accurately – that have been marked as having left the course before 24 November. A lot of students will leave before that date – with all kinds of implications, especially this year.

Back in the day, it used to be before 1 December, as that was the HEFCE deadline for student funding purposes (it still is the HESES census date, which is important for the student funding that comes via OfS). As the 90s became the 00s providers with a wider range of term dates joined the sector, and latterly the funding date has become less important. So the cut-off moved from a date to a time period that would be close enough to the old rule in most cases.

The main HESA Student record uses a different rule – it includes only students that remain beyond two weeks after their commencement date. If you are wondering when that is:

If induction periods immediately precede the start of teaching, the [commencement date] should be the start of the induction period. If there is a significant gap between the two, the [commencement date] should be the start of teaching.”

Lots of potential for confusion there – but I should add that the commencement date is not necessarily the same as the “first day of the first academic year” that the Student Loans Company use. But more of that to follow.

How many students left your provider before those 50 days were up? We get published totals of all (FT UG UK-domiciled first degree) students during their first year from the HESA student record and a published total of all (FT UG UK-domiciled first degree) students who stayed for more than 50 days from the UK key performance indicator. The latest available year of data for the latter is 2017-18, so some simple arithmetic gives us all the students who left after two weeks but within 50 days, with some other modifications based on the HESA-HESES reconciliation process following the closure of the collection – all rounded to the nearest 5 in true HESA style.

[Full screen]

This is the closest we can get to seeing this hidden attrition. It feels like the kind of thing we need to know more about – in any year, but particularly in 2020.

Cooling off

There’s another time limit that universities will be acutely aware of. As with any other contract, starting on a course with a provider has a cooling off period – during which the contract can be cancelled without penalty. The duration of this period varies around the sector – anything between two weeks and one month after the start date. If a student leaves their course during that period – and informs their provider that they are doing so – their finance agreement will be cancelled. So – for a student registered with the Student Loans Company – maintenance loans will never be sent to the students, fee loans will never be sent to the provider, and the student will have nothing to repay.

When does that grace period start. Again the guidance is not as clear as it could be:

[Payment can only be received] once the learner has been in attendance for at least two weeks. This will either be two weeks from the learning aim start date or two weeks after the learner has started their studies (when they start at a later date); this is referred to as the ‘initial liability point’.”

With such a painful start to the 2020 autumn term, there is a worry that many students will never have reached their “initial liability point”. This lowers the expected income of the university based on what was predicted before students started their course – prospective students who either never started or withdrew within the cooling off period will not have fees paid for them. There is no way to see this accurately in the public data nationally – for this year or any other.

I’ve tried matching SLC data by provider and domicile for full-time undergraduate students with similar data from HESA for 2017-18 (the latest available from SLC). The differences between the two values could relate to students who don’t take up SLC fee loans, or students who leave before the fee is paid to the provider, or both. Or possibly something else I’m not aware of.

[full screen]

There’s been some talk of a “cut-off” date in term one, beyond which a student would be liable for term two (25 per cent) as well as term one (25 per cent) fees. Though an arrangement such as this may apply to accommodation contracts (for example), the SLC guidance is clear that there is generally one attendance confirmation deadline per term – the Thursday before each payment date (third Wednesday in October, first Wednesday in February, and first Wednesday in May).

These confirmations should be accurate and timely – though for this year only (2020/21) it is possible to confirm term 2 attendance at the same time as term 1. This is a fix to allow for the payment of term 2 fees to providers in term 1 – student accounts will still show only the first 25 per cent of their fee for term one, until attendance is confirmed actually in term 2. If a student doesn’t turn up in term two a provider is required to submit a change of circumstances as soon as possible – funding received by the provider in term 1 with respect to term 2 will be clawed back.

What about enrollment? That’s a provider-level thing, which sometimes has a bearing on the cooling off process. A student may feel like they had joined the university at the date they got their library card, IT log in, and joined their students’ union (unless they are one of about five annually that choose not to) – but this is not necessarily the “start date” or the “commencement date”.

What we know now

There are some students who leave their course very early in term one – exactly at this time of year. There’s a noted peak within six weeks in some providers – better data would hugely increase our understanding of this. We don’t know and never will know how many students do this with any degree of accuracy for this or any other year.

There is also another pool of students who simply never show up, despite having been placed via UCAS. Again, there’s no reliable way of knowing even vaguely how many there might be in this category.

It feels like a big gap in what we know about students, and potentially a huge hole through which potential and aptitude is being lost. There may be implications for eligibility for further study funding, even when a student lasted just a hair over two weeks beyond induction – this feels unfair generally, but particularly so given what first year undergraduate students have had to endure this year.

And information we can’t really see might mean the difference between a financially sustainable provider and one that is not sustainable. Providers may already know their own position – but a government that is at least nominally concerned about the local and regional HE offer most likely does not. I’d argue that this knowledge is now more important than ever.

Leave a Reply