Category Archives: Data

The phenomenon of university undermatch, and why getting poor students into university just isn’t enough

By Gill Wyness and Lindsey Macmillan

Higher education has long been thought of as a tool to equalise opportunities, with governments around the world spend billions per year on encouraging disadvantaged students into university through financial aid and other widening participation strategies. Indeed, the Office for Students has recently set ambitious new targets to encourage universities to widen access. But is simply getting poor students into university enough? Our new research project, funded by the Nuffield foundation*, suggests that we need to pay much more attention to the types of universities and subjects that disadvantaged students enrol in, if we really want to improve their life chances.

As a team (along with co-authors Stuart Campbell (UCL Institute of Education), and Richard Murphy from University of Texas at Austin), examine the quality match between students and the courses they attend, using data on a cohort of students who left school and enrolled in university in 2008. We are interested in whether certain groups (e.g. disadvantaged students) are more likely to undermatch, by attending courses that are less selective than might be expected given their A-level grades. We also examine whether certain types of students overmatch – i.e. attend courses that are more selective than might be expected given their grades.

We examine this phenomenon of mismatch along two dimensions of course ‘quality’. First, we consider a student to be well matched to their course if they have similar A level scores to others on the course (attainment match). For example, a high-attaining student would be well matched if they attend a course with equally high attaining students. They would be under-matched if they attend a course where their fellow students have lower grades than they do (suggesting they could have attended a more academically prestigious course), and over-matched if they attend a course where the other students on their course have higher grades than they do.

Second, we rank courses based on the average earnings of their graduates 5 years later, and consider a student to be well matched if that course has a similar ranking to their own individual ranking by attainment (earnings match).  For example, a high attaining student would be well-matched if they attend a course with high earnings potential, and would be under-matched if they are high attaining, but their course has low average earnings.

We find a significant amount of mismatch in the English system, with around 15-23% of students under-matching and a similar proportion over-matching. Importantly, we find that students from low socio-economic status (SES) backgrounds are more likely to undermatch than those from rich backgrounds. Comparing low and high SES students at every level of attainment, disadvantaged students attend less academically prestigious courses, and courses with lower earnings potential, than those from high SES backgrounds. So these students have the same A-level attainment, but are attending lower ‘quality’ courses. This has obvious implications for equity, and for equalising opportunities.

But economic disadvantage is not the only dimension of inequality we study. Examining mismatch by gender, we find that female students attend courses that are just as academically selective as male students (attainment match), but they attend courses which have lower future average earnings than men, comparing students with the same A level attainment. This has important implications for equity and for the gender pay gap.

So what should policy makers do? We examine three important factors which might drive this mismatch in an attempt to work out potential policy solutions. First, we consider the choice of subject studied at degree level, comparing students of similar academic attainment and studying the same degree subject, the gap between advantaged and disadvantaged students remains. This tells us that low SES students are studying at lower ‘quality’ institutions relative to high SES students, rather than choosing lower ‘quality’ subjects for their courses.

What about the role of geography? It is well known that low SES students are more likely to attend universities close to home, but does this drive them to choose a less selective institution? If we just consider the group of students living close to home, we still see differences in the institutions that disadvantaged students attend compared to more advantaged students. High attaining low SES students tend to enrol in post 1992 institutions near home, whereas high attaining high SES students are more likely to attend a nearby Russell Group university. There may therefore be scope for some outreach work for high ranking universities to attract local disadvantaged students. Interestingly, those low SES students who move further away from home to attend university appear to be as well-matched as similar attaining high SES students.

Our third factor of interest is school attended, which we find accounts for the majority of mismatch among low SES students. The implication is that factors correlated with high school such as peers, school resources, information, advice and guidance (IAG) at school, and sorting into different types of schools, play an important role in student match. Unpicking what it is that is driving this important schools channel is an important step for future research.

Turning to our gender gap in earnings mismatch, we find no role for distance to university or schools attended. But we find a very important role for degree subject studied. The fact that women attend courses with lower future average earnings than men is largely driven by the subjects that women are studying, rather than the institutions they attend. For example a high attaining male student might choose a subject such as engineering, which is typically high returns, whereas a high attaining female student might choose a subject such as English or History, commanding a lower average salary.

So what can we do? The evidence suggests that an intervention that may help to reduce SES and gender gaps in match would be to improve the level and quality of information available to under-matched students, for example on the attainment profile of students on each course, and labour market returns.

Some recent studies have investigated the importance of providing information to low SES students specifically to improve match (Dynarski et al, 2018, Sanders et al., 2018). Our results highlight that it may also be beneficial to target women in a similar way, providing information on potential earnings associated with both institution and field of study. However, as with most studies of mismatch, we have no information on the preferences of students. Women may be well-informed on the earnings potential of subjects, but simply prefer not to study them. Similarly, it may be the case that low SES students prefer to attend less academically challenging institutions even when their attainment levels suggest they are academically prepared. This could be down to perceptions about institutions not being a good fit for them. Our finding on geography suggests that university widening participation units could do some important outreach work in these cases to challenge perceptions (Sanders et al., 2018).

read more in our working paper: http://cep.lse.ac.uk/pubs/download/dp1647.pdf

this blog first appeared on wonkhe, on 5th December 2019

*The Nuffield Foundation is an endowed charitable trust that aims to improve social wellbeing in the widest sense. It funds research and innovation in education and social policy and also works to build capacity in education, science and social science research. The Nuffield Foundation has funded this project (172585), but the views expressed are those of the authors and not necessarily those of the Foundation. More information is available at http://www.nuffieldfoundation.org.

Small Numbers and Strong Statements: Analysing offers to black students at Oxford

By Jack Blundell

Under immense pressure over transparency, Oxford University last month released its latest and most-detailed-ever tranche of undergraduate admissions data. These statistics have been heavily drawn on in a series of articles and statements, notably by the MP for Tottenham David Lammy (Link) who focuses on the representation of black British students. The goal of this post is to help inform this particular debate by analysing relevant features of this treasure-trove of data, pointing out where we can infer true differences from patterns in the data. Unless otherwise noted, the source of all analyses below is Oxford’s May 2018 Annual Admissions Statistical Report and the associated datasets released.

To summarise my discussion, I find that:

  • Among Oxford offer holders, black students are under-represented relative to the overall population but over-represented once we restrict to those achieving the top school grades.
  • Out of those who apply, black students have a significantly lower probability of receiving an offer.
  • Just under half of this gap can be attributed to black students being more likely to apply to more-competitive subjects than other applicants.
  • The remainder of the gap is due to lower probabilities of application success within courses.

Continue reading →

The relationship between A-level subject choice and league table score of university attended: the ‘facilitating’, the ‘less suitable’, and the counter-intuitive

By Catherine Dilnot, UCL Institute of Education

As the school exam season gets under way, English 18-year-olds hoping to go to a selective university will typically be taking papers in only three A-level subjects, chosen two years earlier from scores of possible subjects approved nationally, although in practice from the somewhat smaller number offered by their school or 16-18 college.  This early specialism in so few subjects can have long-term consequences.

For many UK degree courses particular A-levels will be required – for example biology and chemistry for medicine.  But many others don’t have subject pre-requisites, including popular degrees like business and law.  So whether a sixteen year old isn’t yet sure what they want to do at university, or has an idea but wants to do a course without pre-requisites, it’s difficult for them to know which subjects to choose.  The question then is whether some of the large number of A-level subjects available are more helpful than others in getting them to the university of their choice.  Recent reforms have reduced the number of A-level courses approved for teaching in English schools from over 90 to 60, but it is still a bewildering array, both for students choosing, and for schools and colleges deciding what subset to provide.

One important reason that subject choice matters is because we know the sorts of A-levels chosen by 16-year-olds vary by socio-economic background.  And while the number of young people going to highly selective university from low SES backgrounds has increased over recent years,  UCAS figures for 2017 show that an 18-year-old in the top SES quintile is ten times as likely to attend than someone at the bottom.  It’s clear that most of this gap is a result of differential prior attainment, but evidence on whether some subjects are helpful for entry to highly selective university could help chip away at the SES gap.

Continue reading →

Grade prediction system means the brightest, poorest students can miss out on top university places

By Gill Wyness

With UK tuition fees now among the highest in the world, but benefits from having a degree remaining substantial, choosing the right university has never been more important for young people. The government has tried to make this easier by offering more and more information not just on the university experience but on the quality of the institution and even the potential wage return students could reap.

Despite all these efforts to make the decision about where to apply as informed as possible, one issue remains: students still apply to university based on their predicted rather than actual qualifications. And these predictions are not always accurate.

Using information on university applicants’ actual and predicted grades and their university attended, obtained from the Universities and Colleges Admissions Service (UCAS), I find only 16% of applicants achieved the A-level grades that they were predicted to achieve, based on their best 3 A-levels.

Continue reading →

Ethnicity trumps school background as a predictor of admission to elite UK universities

by Kurien Parel and Vikki Boliver (University of Durham)

 Last year an article in the Guardian newspaper described significant disparities in the success rates of white and non-white applicants to the University of Oxford, even among students who received top grades at A-level.  The article reported that, in 2010-11, offer rates were around 1.5 times higher for white applicants than for ethnic minority applicants with the same grades, and up to twice as high in relation to Oxford’s two most oversubscribed subjects, Medicine, and Economics and Management. This pattern was found to hold even for students with 3+ A* grades at A-level. Continue reading →