Comment Article 19 January, 2017

Lessons from today’s education data release

This week the Office for Budget Responsibility (OBR) forecast that total spending on education, as a share of GDP, is to drop from 4.4 per cent currently, to 3.9 per cent by 2022. This is a serious issue given that investing in education can prevent spending even more later (for instance on benefits and healthcare), as well as increase GDP and tax receipts.

Given the realities of the constrained current context, making smarter use of data is a key strategy for improving outcomes while still making efficiencies. It can lead to better quality public services that are more personalised and more responsive. It can save costs by increasing speed, decreasing duplication and automating some functions. It can also lead to greater control and choice for citizens, leading to focused accountability of providers, and so driving up quality.

But such change is hard in public services, with departmental siloes and strong systemic resistance to any potential risks (and data protection/security often used as reasons for inaction). However, the political willpower and leadership can be found in places. For instance just yesterday John Manzoni, the Chief Executive of the Civil Service, expressed his desire to see data and analytics become the “life-blood of operational decision-making in government”.

Therefore, within the deluge of education data released openly today, which should itself be applauded and accelerated, two particular reforms should be highlighted as examples of good practice that other sectors could learn from.

The LEO dataset of destinations, linking education to earnings

A quick browse on scholarly search engines shows just how much the impact of education on earnings has captivated  the minds of researchers. Today, the Department for Education (DfE) is publishing its annual release of KS3/KS4 results, school league tables and destination data. Although it wasn’t updated today, it’s still worth focusing on the ‘experimental’ Longitudinal Education Outcomes (LEO) dataset, first released August 2016.

It was only in 2015 that the Small Business and Enterprise Act allowed for the linkage of higher education, employment and earnings data 1, 3, 5 and 10 years after graduation. This is the holy grail for anyone interested in understanding the impact of education on later outcomes in life. Over time it will help policy-makers and providers to understand which aspects of education (such as particular higher education institutions (HEIs), courses or qualifications) can lead to better outcomes for different learners.

Importantly, by linking existing datasets it does not impose any additional data collection burdens or costs on universities, employers or members of the public. It also shows impressive cross-government collaboration and commitment between DfE, DWP and HMRC (and formerly BIS too).

In the future, such data-linking approaches could be extended to the whole education and tax system, including the Individualised Learner Record (ILR) dataset for colleges, student loans. By linking to health, welfare and justice data, it could also start to look at a wider range of outcomes for people. The hope must be that John Manzoni’s desire for a data revolution in government will pave the way to this promising future.

The Progress 8 accountability measure for schools

Introduced as part of Michael Gove’s accountability reforms, Progress 8 is a new measure for KS4 pupils which is being used to hold schools to account for the first time this year. It replaces the previous measure – the proportion of pupils achieving 5 A*-C including English and Maths – as the new floor standard. Progress 8 measures pupil performance across eight core (mainly more ‘academic’) subjects, compared to the performance of pupils with similar prior attainment at KS2.

There are three main aims to the new measure. Firstly, to incentivise schools to focus on all pupils equally and not to simply prioritise the C/D borderline pupils. Secondly, to call ‘coasting’ schools to account, those with a high prior attaining intake of students, who are more likely to also attain highly at KS4. Thirdly, to incentivise schools to get more pupils taking EBacc (i.e. more ‘academic’) subjects, which are shown to improve later employment outcomes – an aim that appears to be working given the proportion of students entering EBacc subjects has increased by 0.6 of a percentage point to 36.8%.

Progress 8 is one of a raft of accountability changes which serve to reduce perverse incentives, such as the early-entry rules and updated qualifications. Despite the challenges communicating such technical reforms, in general more robust outcome-focused accountability measures such as these are to be welcomed, and there should be more of them across public services.

Working across different public services gives Reform a privileged ‘big picture’ perspective.  These two successful reforms led by DfE are valuable case studies that other Departments could learn from.

This article was co-authored by Louis Coiffait, Head of Education, Eleonora Harwich, Researcher and Elaine Fischer, Research Assistant at Reform