Unexpected peaks: widening participation in Scotland

The Scottish government’s recent emphasis on improving recruitment from the most disadvantaged postcodes is getting results. But, as ever, it pays to dig a bit deeper.

The end-of-clearing figures produced by UCAS show the participation rate at age 18 for Scots from most disadvantaged 20% of postcodes (SIMD 1) now stands at 12.9%. Last year it was 12.1% and as recently as 2015, just 9.8%. Entrants from SIMD1 have risen this year from 1,410 to 1,470 (+4%), at a time when the total number of 18 year olds is falling and the number of acceptances from all other SIMD quintiles at that age has either stayed the same or dropped (see Table1 below).

Background and targets

This change has been achieved without additional investment on any scale, or any direct financial incentives. An irresistible force has instead been applied in the grey area between moral pressure, individual universities’ desire for political favour, the vague threat of financial penalties, and more general fear over funding for the sector should the government’s targets on widening access not be met.

The first of these targets requires that by 2021, 16% of total Scottish university entrants should be from the 20% most disadvantaged areas, and that these should represent at least 10% of Scottish entrants in any individual institution. In the current all-age UCAS figures, SIMD1 acceptances are 15% of the total.

The mysterious increase in older entrants

The recruitment of Scottish and EU students is tightly controlled, as is common in systems where the government cash budget is meeting the whole cost of teaching. Very few new places were put in this year. The government might therefore have been expected to express some surprise when UCAS revealed around 500 more entrants than last year, when looking across all age groups. Yet the government lines to the press celebrate “a record at this point in the year”: the lack of a formal press release on the government site is the only sign of caution.

Some caution is wise, as this overall increase in entrants is probably a trick of the light. Surprisingly, it is due substantially to an increase in students from the higher SIMD quintiles, aged 19 and more.

Table 1: Scotland: Increase in acceptances by age and SIMD: change 2017 to 2018 Day 28 of clearing

SIMDAge 18Not Aged 18 
160100
2-1050
30100
4-10100
5-90170
Total-50520
Age
17 and under10
18-30
19230
20-2470
25 and over230
Total510

Note: Totals do not match exactly due to cases with missing data.

It seems likely that a course or group of courses which has been moving large numbers of people from college-level HN to degree level study for years has suddenly started to log these moves through UCAS. The increases are concentrated in four areas (professions allied to medicine, biological sciences, computer sciences, and business and administrative studies) the last two of which at least are strongly associated with college to university articulation.

The Scottish Funding Council has been encouraging institutions to make more use of UCAS to process articulating students. This is good for the comprehensiveness of UCAS data, but not so good for reliable time series data. It strongly suggests some of the recent growth in entry for students over 18 recorded through UCAS, not just this year but potentially over the past few years, has been due to a decline in the number of articulating students who bypass UCAS.

The likelihood that more articulation is now going through UCAS’s books also raises a question about the standard disclaimer that UCAS figures omit many Scottish students who enter HE through the college system: that will be true at 18, but less so once those over 18 are included.

The real increase at age 18

Luckily, the figures for 18 year olds, mainly school leavers, are not so susceptible to administrative effects. This is where we can be most sure we are seeing the effect of the Scottish Government’s emphasis on giving priority to people living in SIMD1 areas.

Universities have devised various schemes, including ones related to admissions criteria, greater access to clearing, grants, and free accommodation, for which a SIMD1 postcode can be a necessary and sometimes even a sufficient condition. SIMD1-based schemes began to appear last year, but have been more visible in this round. I doubt it is coincidental that this is also the first year I’ve been told a story about a wealthy family gaming the system, by using an alternative address on the admission form. It isn’t clear how thoroughly institutions check the addresses on UCAS forms for plausibility: up to now there has been little to gain being creative about describing where you live. As that changes, it may be time to look harder.

However, this is not to suggest that the rise from SIMD1 areas is down to a sudden surge of sharp elbows. As David Kernohan has already noticed, an even larger rise is visible in the figures for areas with very low participation in HE (POLAR4 Q1). For technical reasons, Scotland is underrepresented in this data set. So the absolute number of Scottish cases is small in the UK bottom quintile: the 12% rise is from 620 to 690. As there is no distorting effect on the POLAR figures from being a target, they provide some reassurance, I think, that we really are seeing increased entry from particularly underrepresented areas, and not just, say, a shuffling of the better-off people in SIMD1 areas from college to university.

Disadvantage upon disadvantage outside SIMD1

The news looks less good for disadvantaged people living outside SIMD1. The SIMD2 figures are the best canary down the mine we have for the situation of more disadvantaged young people lacking a SIMD1 postcode. If, in a tightly capped system, the increase from SIMD1 was restricting opportunities for other groups, it is those likely to have less strong results and therefore the areas with the next-highest concentration of more disadvantaged people, which you would expect to be most affected. And this is exactly what the numbers show.

At age 18, people from SIMD2 areas are seeing a small drop in number, but they are not alone in that. More seriously the participation rate for SIMD2 is at a standstill, uniquely among all groups.

Table 2: Year on year change in age 18 participation rates by SIMD1 and POLAR4 quintilea

 2017 SIMD2018 SIMD2017 POLAR42018 POLAR4
112.10%12.90%10.80%12.10%
218.30%18.40%15.40%16.50%
323.70%24.50%20.60%20.00%
430.60%31.70%26.20%26.50%
540.70%41.90%37.70%39.20%

Until very recently, the growth rates in acceptances from SIMD1 and SIMD2 were very similar. It is only in the last two years that the numbers have pulled so far apart (figure 1). Whether those from SIMD2 areas are getting less encouragement to apply, are becoming relatively less likely to do so in the absence of eligibility for various schemes, or are less successful when they do, because less is done to put their exam results into any wider context, can’t be understood from these numbers, although these are all potential explanatory effects.

A similar pattern emerges in the POLAR4 data: acceptances from the bottom two POLAR quintiles, which together account for around 15% of acceptances, have increased, but at the next step up, POLAR4 Q3, which accounts for the next 15% of areas, there has been a fall of 6% and, uniquely, the age 18 entry rate for this group has actually dropped this year (table 2).

In 2016, 40% of children in “families with limited resources”, as defined by Scottish Government researchers, had a SIMD1 postcode, another 29% had one in SIMD2. Thus a large minority of the most disadvantaged young people live in SIMD2 areas. The pressure on universities to recruit more students from the most disadvantaged postcodes is working on its own terms. But in the absence of any headroom for expansion, and even with a falling number of 18 year olds, disadvantaged people who live on the wrong side of the tracks are becoming squeezed.

The wider context

Still, here just for once is a Scottish story which is not all about free tuition. Except of course that it is. The heavy pressure being felt now to widen access is rooted in the long period from 2007 onwards when the numbers hardly shifted. It turned out that 100% universal free tuition did not, if left to itself, make much difference, an increasingly uncomfortable political situation as time passed. Ministers these days talk about free tuition being a necessary but not sufficient condition for widening access and point to the new activity they are stimulating. Whether the argument about free tuition’s essential contribution is convincing is neither here nor there: the political storytelling about free tuition, and a little bit of the First Minister’s reputation, now rely on the targets being met. Hence the heat universities are feeling.

If Scotland offers a lesson in how to shift the behaviour of a higher education system at speed without resorting to bribes, it is that keeping institutions in a state of anxiety about their financial future and in competition for political favour can be very effective. There are also echoes of 2013, when student grants were cut while spending on tuition fees was protected. Now, the evidence points to disadvantaged young people outside SIMD1 areas being squeezed most, even while the SIMD1 figures are being successfully driven upwards. Yet again in Scottish HE, the rhetorical investment in free tuition is casting its long shadow, and being in or out of the political spotlight makes all the difference.

2 responses to “Unexpected peaks: widening participation in Scotland

  1. Interesting analysis here, with the little addition that contrary to the assertion the government has not always met the whole cost of teaching!

  2. A further thought. Universities use a wider range of contextual admissions data than just SIMD, which is known to have drawbacks. The quintile 2 POLAR data is positive (though this is not commented on in the same way that POLAR for Q1 is highlighted as evidence that the progress is genuine for that group) and it may well be that as the political focus has shifted from MD40 to MD20 universities are taking the opportunity to look at a wider range of contextual factors for students not in Q1, which is reflected in the POLAR but not the SIMD data.

Leave a Reply