Let me take you back to March 2017, and a report from the Behavioural Insights Team.
The consultancy spun out of David Cameron’s infamous “nudge” unit conducted a study whereby students with good GCSE results attending a school that sent more than 20 of high achieving A level students to their nearest higher education provider were sent letters ostensibly from Russell Group students to encourage applications to and acceptances from places like Bristol instead.
Long story short, if you write a prospective student a letter (or two letters) suggesting they do something, they are a little more likely to do it. Not much more likely (only a letter to home and school breaches even the p<0.05 barrier for applying to and accepting a place at a Russell Group provider) but there is an impact.
Now, if you are anything like me you’ve got ethical alarm bells ringing all over the place. How would the OfS react if a university chose to write a faux-student letter to every promising A level candidate in the country declaiming the delights of study within their cloisters? How many of the students in the study dropped out of a Russell Group university finding that this particular experience was not for them?
Back to the future
Why bring this up now? Well, something called “Clearing Plus” has emerged as a part of what people are still somehow calling the sector Covid-19 bailot. As explained by DfEs, it is:
a new service which matches students to universities or other opportunities based on their achievements and course interests. If students’ calculated grades exceed their predicted ones, it can suggest alternative courses with higher entry requirements.”
The old education technology sceptic in me insists that we also note that this is the main plan of a “new, personalised, Clearing system”. In essence, unplaced applicants (those who will definitely enter clearing) will sign in to UCAS and see a list of courses they have been “matched” to, with easy links to send an expression of interest.
If someone with a conditional offer already firmed up fancies even a look at what they could have won, this list of matches becomes available as soon as you break up with your existing choice (yes, it is all a little bit like Tinder) and enter clearing via Self Release.
As UCAS’s Clare Marchant describes it:
We are confident the new personalised Clearing in 2020 will transform the experience for students. They won’t need to search through a mountain of courses or make endless phone calls. The most appropriate course options for them as an individual will be presented through their online account.
Clearing Plus builds on the successful introduction of the online self-release last year, to give students more control while also making the process smarter, fairer, and more personal. Our initial projection is that we expect around 50,000 students to use Clearing Plus to discover their course.”
A difference in emphasis
Fifty thousand students is a hair under 10 per cent of all last year’s accepted applicants – equivalent to the number who accepted a place via main scheme (rather than direct) clearing. It’s double the number who changed course via self release last year. The wording of the information we have so far suggest that this projection includes a proportion of those expected to be placed via main scheme and direct clearing, including those who self release.
What’s striking is that Clearing Plus is emphatically not a way to get applicants with firm offers to ditch them by showing them what else is on offer – applicants would have to ditch their firm offer to see the alternatives. It’s a replacement for Direct Contact, which saw providers contact students who might be interested in courses directly. To be clear, it is not a planning mechanism that seeks to move applicants with better grades up to a higher class of provider. At least, not based on what we know so far.
And yet – if you were just to read the information provided by DfE that is what you would conclude. The sector support offer suggests something that intervenes more directly – you could imagine something as described by UCAS that offered these choices to all students who did better than expected. Or, more insidiously, merely informed an applicant that non-specific “other choices” would be available should they opt to throw over their existing plans. Not now, but merely not yet?
It is, emphatically, a way of influencing student choice by presenting options they may not have considered. The question of how these options are generated is something that we need to seek very detailed answers to.
Under the bonnet
Personalisation – I suspect – is a key selling point. But the impression this of a pulsating electronic brain telling applicants where to study is – as with online shopping recommendations – likely overplayed. The algorithm is likely to impute someone’s subject interests from their mainstream applications and then match the actual grades against required grades for courses not considered to identify matches. Keeping it as simple and transparent as possible.
That said, I’d hope it would include other preferences too – the range of reasons a student chooses a course are many. For an applicant interested in politics, not feeling you would get A*,A,A (with one of the A grades in mathematics) may not be the reason you had not selected a BSc (Hons) in Economics and Politics at the University of Bath (A perfectly decent course at a perfectly decent university, if that isn’t clear).
Maybe you had considered Bath and rejected the idea. Maybe you live in Sunderland, and – while you like the idea of leaving the area you grew up – you don’t want to be that far away from your friends and family. Maybe there’s a particular specialism at your first choice university that you’d set your heart on, and that Bath doesn’t offer. Maybe you feel like Bath as a city is not the place for you. Maybe your boyfriend is going to Lancaster. Maybe your sister is already in Bath. Maybe you just like the idea of studying at the place you’ve been dreaming about since you filled your form in over the Christmas break.
All reasons – good, bad, and bizarre – that applicants use to decide where they study. Worth stating, because sometimes we lose sight of the personal and idiosyncratic nature of the life choices 18 year olds make as we seek to use them to direct the proceeds of multiple billions of pounds of government borrowing each year.
Seems to be some creative representation of old/established services here. Direct Contact Service with Adjustment (no doubt encouraged). Sure the interface and process will be different, but I can’t see anything new here, though I might be missing something.
I am confused as to why you’d build this service on a set of wholly untested grades (the new awarded grade), when UCAS has full access to yeas of predicated grade data which are more reliable.
When a university takes student via the service is the CAP fee the standard or are they charged a premium as per DCS?
Seen something which says the service is free to use, but in the detail does this mean there is no placement charge?
I can’t see anything very “New” in Clearing Plus. Looks like DCS with a hope of Adjustment and Selr release in over-drive.
I’m sure any interface and process will be different, but not sure it is new.
Wonder also why use new awarded grades rather than predicted grades to drive the model. The new grades are untested and therefore less reliable than the well established predicted grades, and UCAS hold years of that data.
Lastly, what’s the placement charge? A standard CAP or an enhance charge?
Not at UCAS anymore but I do remember doing the analysis on DCS and speaking to customers about it.
The number one item on their list was an easier interface, there was to much emphasis on the ‘cleverness’ of the algorithm. And that just caused no ends of issues. So it’s great to move in the direction the customers wanted and need.
Good Luck, it’s a hard year for everyone so I hope this supports universities and students a like.
I would like to understand further how the system selects the universities for the young person? Aware there is an algorithm, can this be explained in layman’s terms? I would be fearful that not all universities would be represented equally.
This seems very similar to a service that UCAS was expecting unis to pay to be featured on/promoted by, given a light dust off. Presumably without the ‘pay to play’ element, or not?
Nudge related admissions interventions are under a cloud as a consequence of the Harvard admissions class action. To rewind, Harvard sent letters to students doing well in SATs but the score needed to trigger the letters varied depending on geographic, socio, and ethnic characteristics. (Also SATs are under a cloud as well for all the PQA fans out there – but that’s another story). A slew of studies conclude that cheap interventions are not scaleable unless you toss in actual resource – a financial incentive. Bespoke nudges do have a (small but significant) impact but only because they are given by members of a group lready known to / respected by the recipients. Which is another way of saying, of course, that real resource is needed.
As for the ethnical alarm bells, it is worth pointng out that students live in a world awash with recommender algorithms (Youtube, Facebook…). Clearing+ is pretty benign in comparison.
There’s no way Clearing Plus will get that many students using it. It ONLY produces matches on applicants with in year results from ABL (so all those Direct clearing applicants wont be matched, gap year and reapplicants wont me matched etc). Plus it’s really limited in terms of which ABL qualifications it will match on (A levels only, BTEC Extended diploma only – no combinations or international qualifications at all and they were struggling with Access last time I checked). This is a tool they’ve developed to help a standard A level 18 yr old clearing applicant from 2011 – it isn’t fit for purpose in 2020.
They’re were still offering providers an “enhanced” option for a price – that might be off the table now that they’ve promised the government that this is an amazing improvement