Education and the 2019 general election

16th December 2019 by Timo Hannay [link]

What does the recent UK election result mean for education? The truth is that no one outside government (and perhaps even inside it) yet knows. But here we take a look at what education – along with various other socioeconomic and demographic indicators – appeared to mean for the election, at least in England. We find that:

• Despite gaining many traditionally Labour-supporting seats, the Conservatives continue to represent parts of England with lower average levels of childhood poverty, higher proportions of young people in education and employment, and higher average levels of life satisfaction among local populations.
• However, when we look at the swing from Labour to the Conservatives, a different picture emerges. Areas with the largest swings tended to have high levels of childhood poverty, low academic attainment at secondary school and beyond, and low levels of participation in higher education. Broadly similar patterns were evident for Brexit Party support where they fielded candidates.
• Statistically speaking, then, the Conservatives continue to be the party of more affluent and contented parts of England. But the surge in support that they enjoyed during last week's election came largely from apparently 'left behind' areas with much lower levels of affluence and educational success.

Lessons learned

Figure 1 below is an updated version of this figure from our previous study on the relationships between educational, social and political factors across England. It leaves the original data unchanged, but adds results from last week's election so that we can look at any new correlations that may have emerged. Each dot represents a local authority area (not a Parliamentary constituency). Dot sizes correspond to pupil numbers. Hover over each dot to see accompanying data; click on the figure legend to turn individual regions on or off. (See Footnote 1 for details).

Let's first examine the factors that seemed to affect Conservative vote share. Needless to say, this negatively correlated with Labour vote share (with a few exceptions in areas, such as Kingston upon Thames and Richmond upon Thames, where the Liberal Democrats did well) and positively correlated with support for Brexit during the 2016 referendum. Despite the Conservatives gaining many traditionally Labour seats, there is still a strong tendency for more affluent areas to show higher Conservative support. There is a similar pattern when we look at pupils who are eligible for free school meals or for whom English is not their first language. The Conservatives also received more support in areas with lower per-pupil funding (which generally correspond to non-metropolitan areas) and lower primary school attainment for disadvantaged pupils but higher proportions of students staying in education or employment at age 16. Their share of the vote also positively correlated with population wellbeing measures such as life satisfaction.

Support for Labour shows mostly the opposite trends to those shown above, so we won't go into them in detail here and will look instead at support for the Brexit Party. This analysis is complicated by the fact that they did not contest Conservative-held seats, but we can nevertheless see clear trends by poverty, low house prices (especially clear outside London), low social mobility, low participation in higher education and, of course, high support for Brexit in the 2016 referendum. Among schools in Brexit Party-supporting areas, we see higher proportions of pupils eligible for free school meal, high levels of pupil absence and persistent absence, and low proportions of pupils with for whom English is not their first language. There are also relatively high average levels of socioeconomic segregation between schools and low self-generated school income. When it comes to educational performance, schools in Brexit Party-supporting areas show lower average attainment at Key Stage 1 and GCSEs, as well as lower proportions of students achieving Level 2 and Level 3 qualifications.

Finally, we will look at one of the key characteristics of this election: the Labour-Conservative swing. Unsurprisingly, this correlated positively with support for Brexit in the 2016 referendum. It also correlated positively with childhood poverty, pupil absence, socioeconomic segregation among schools. It correlated negatively with GCSE attainment, proportions of students achieving Level 2 or Level 3 qualifications, participation in higher education and the proportion of the local population with degrees.

Use the menus above Figure 1 to explore other relationships. See also our previous analyses of socioeconomic and demographic trends in education (Part 1, Part 2 and Part 3) for an exploration of the effects of poverty, urbanisation, Brexit and more.

One nation?

Overall, these results suggest that the Conservatives continue to be the party of the relatively affluent and contented parts of England. But the surge in support that they enjoyed during last week's election came largely from areas commonly thought of as 'left-behind': those with higher poverty and socioeconomic segregation, but lower educational attainment. The challenge in education policy and elsewhere will be to justify the votes of these new supporters without alienating their traditional base.

As ever, we welcome your thoughts: [email protected]

Footnotes:

1. Results from each constituency were compiled from BBC News. In order to enable comparisons with the other data sets in this analysis, each constituency was then associated with its corresponding local authority (LA) area, and average election results for each LA calculated. Most constituencies lie within one and only one LA, but where this was not the case it was associated with the LA that is linked to the largest number of schools in the constituency. Chorley was omitted from this analysis because its Member of Parliament is the Speaker of the House of Commons and neither the Conservatives nor Labour fielded candidates there.

MARK II: What we learned from over three million primary test results

12th December 2019 by Timo Hannay [link]

Last year we were lucky enough to publish a preliminary analysis of an exciting new data set in primary education. It was the result of a collaboration with the team at RS Assessment from Hodder Education. Their amazing MARK database gathers assessment test results from thousands of school across England (and beyond), providing each school with a convenient way to monitor their own results online.

In addition, by analysing the data in anonymous, aggregate form, MARK can also provide a unique perspective on what's happening in primary education across the country. To that end, we have now updated our earlier work with an even bigger analysis of over 3 million tests taken up to the end of July 2019.

This time the data cover grammar, punctuation and spelling (GPS) as well as reading and maths. Once again, we look at the effects of gender, summer birth, deprivation and Ofsted ratings. We also examine whether some schools do better than others at closing attainment gaps between boys and girls, or between summer-born pupils and the rest.

For the answers to these and other questions, see our joint report. For the more curious and adventurous, we have also released a series of interactive charts that go beyond the static pages of the report.

As always, we welcome your feedback and suggestions: [email protected]