Fact Check: are disadvantaged young people 12 times less likely to go to university?

What I know is that there are about 12 times more people from advantaged, privileged backgrounds going to university than disadvantaged backgrounds and that isn’t good enough for me.

Ed Miliband, Labour party leader, in an interview with Evan Davis on the BBC.

We are fortunate in England to have access to administrative data which links the children in our school system with the students in our university system. This enables us to examine how much more likely those from advantaged backgrounds are to go to university than those from less advantaged backgrounds, and hence to check whether Ed Miliband’s statement holds in England. (It is not clear whether his figures refer to England or the UK; requests for further information have not yet been answered by the Labour Party.)

The data for pupils in England include very rich information on attainment at different points in the education system. But they are less good in terms of information on socio-economic background. We can see whether pupils who attended state-funded schools are eligible for free school meals; but this only allows us to differentiate those from relatively poor backgrounds from other pupils.

To identify those from the most advantaged families, we rely instead on information about the local neighbourhoods in which pupils live, including the proportion of residents with the highest educational qualifications, from the highest social classes, and who are most likely to own their own homes. Again, we can only observe this information for students from state-funded schools. Our analysis combines this geographic information with free school meal eligibility at age 16 to create an index of socio-economic background among state school students.

Using data on all state school pupils in England who took their GCSEs in 2008, we can see how many went to university at age 18 or 19 in 2010-11 or 2011-12. Comparisons between the most and least advantaged in this field often focus on the top 20% versus the bottom 20%. Following this lead, we compared the percentage of state school students in the top 20% of our index of socio-economic status who went to university with the percentage of those from the bottom 20% of our index who went to university. This should give an indication of how much more likely state school students from the most advantaged backgrounds in England are to go to university than state school students from the least advantaged backgrounds.

The figures in this piece are similar to those estimated in previously published work by myself and colleagues which used a wider range of cohorts. Our research has looked at the socio-economic differences in higher education participation overall and at elite universities, and at socio-economic differences in dropout, degree completion and degree class.

Three times more likely

Our analysis showed that around 56% of state school students in England from the top 20% of our index went to university at age 18 or 19, compared to about 19% of those from the bottom 20% of our index. This is a substantial difference; but it suggests that state school students from the most advantaged backgrounds are around three times more likely to go to university than those from the least advantaged backgrounds.

CC BY

Even if we were to compare the 1% most and least advantaged state school students – or the 1% most deprived state school students with private school students – we find that those from the richest backgrounds would be around six times more likely to go to university than those from the poorest backgrounds.

These figures come from before the most recent changes to higher education fees were introduced in 2012. But the evidence to date suggests that, if anything, the gap in higher education participation between those from the richest and poorest backgrounds may have narrowed somewhat since then.

There are two important points to note. First, the benefits of going to university generally accrue only to those who complete their qualifications. They also tend to vary according to where you go to university and what degree class you receive – a third or a first, for example.

Our analysis suggests that there are substantial socio-economic differences in these outcomes as well. For example, state school students from the most well-off 20% of our index of socio-economic background are around nine times more likely to go to a Russell Group institution than those from the least well-off 20% of our index – 14% versus 1.6%.

Even among the relatively selected group who go to university, those from lower socio-economic backgrounds seem to do less well, on average, than those from higher socio-economic backgrounds. For example, using the same administrative data described above, but for slightly earlier cohorts, our analysis suggests that the 20% least well-off students are almost twice as likely to drop-out within two years (19% versus 10%) as the 20% most well-off students. Among those who complete their degrees, those from a lower socio-economic background are around a third less likely to graduate with a first or 2:1 (43% vs. 67%).

Second, a key driver of the socio-economic differences in university participation – and the differences in degree outcomes – is how well students do earlier on in the school system. We know that how well young people do in their GCSEs and A-levels is strongly predictive of whether or not they will go to university. Our analysis suggests the fact that young people from the most and least deprived backgrounds have very different GCSE results can explain almost all of the reason why they have very different university participation rates.

This highlights that substantially reducing or eliminating the socio-economic gap in university participation is likely to require interventions that address the very large differences that exist in attainment earlier in the school system.

Verdict

Our analysis suggests that, among state school students in England, those from the 20% most advantaged backgrounds are around three times more likely to go to university than state school students from the 20% least advantaged backgrounds – a lot less than the 12 times that Ed Miliband suggested. But they are also around nine times more likely to go to a Russell Group institution and around 50% more likely to receive a first or 2:1.

Thus, while the socio-economic gap in university participation is not as large as Ed Miliband suggested – and has narrowed somewhat in recent years – there is still much work to do to level the playing field completely between those from different socio-economic backgrounds.

Review

This analysis shows the very stark difference in the probability of going to university between young people from the most and least advantaged backgrounds. Depending on how one defines “advantaged”, the least privileged are between three and six times less likely to go to university than the most privileged. And the gap is much larger if one only considers elite universities.

One important point is that the gap is mostly explained by results at GCSE. So if we want the gap to be removed, more attention needs to be given to what impedes children from disadvantaged backgrounds from progressing up to age 16 – it is not mainly a question of improving access for 18 or 19-year-olds. This fact check supports the spirit of Ed Miliband’s remarks, but not his actual numbers. It is a great illustration of the use to which the excellent English administrative data can be put by researchers. – Sandra McNally

Claire Crawford receives funding for her research from a range of government departments, research councils, charitable trusts and other organisations, including the Economic and Social Research Council, the Nuffield Foundation, the Department for Education, Universities UK, the Education Endowment Foundation and the Sutton Trust. All of her research is independent and the views expressed in this article are entirely her own.

Sandra McNally has received funding from the Economic and Social Research Council and the government to do independent research, but the views in this article are her own.