DfE: Early identification of young people at risk of poor educational and labour market outcomes: the role of educational institutions

Lead Research Organisation: University of Westminster
Department Name: Westminster Business School

Abstract

This study attempts to answer the following questions:
1. How do things like truancy, coming from a less affluent background, family breakdown and a range of other factors that pupils experience at school; lead to poor educational and labour market (employment and earnings) outcomes?
2. Do we see different impacts (for instance on the likelihood of securing good wages) when similar students attend different post-16 educational institutions, such as Further Education, Sixth Form College or School Sixth Form? Does it make any difference to a young person's prospects if they achieve the same level of qualification in these different institutions; and do we see children from rich and poor backgrounds making very different decisions from age 16 and above?
3. How accurately are we able to predict the employment and earnings outcomes for different students, using all the information on their background, learning and achievement in schools and colleges?

At its heart the project seeks to analyse and assess the educational and labour market pathways followed by the half of young people who do not pursue university level education, and therefore contributes to the government's social mobility agenda; emphasised by David Cameron as a key priority for government in his Oct 2015 conference speech. The research proposed here will be of key interest to government and the Social Mobility Commission charged by the government to address Britain's poor record on social mobility.

When we talk of social mobility, we are interested in the extent to which children born to poorer families can make the journey to high paid jobs and professional careers. More generally, a lack of social mobility is a situation where being born to poverty, riches or somewhere in between, means that you are likely to find yourself in the same position as your parents, no matter how hard you try and whatever your talents. Unfortunately, the evidence over recent decades has been that there is less social mobility in the UK than in other similar countries.

The administrative data we will use to carry out this study is routinely collected by the parts of government that collect taxes (HMRC), deal with unemployment support (DWP), are responsible for Further Education (BIS) and learning in schools (DfE). This is a very important and useful resource, as it has the potential to overcome some of the limitations we face when using surveys (not least that we observe all people in the relevant populations, not just a relatively small sample). However, it is a complicated process to link these datasets and a large part of this project will be taken up with this process of linking.

As well as finding out what impact truancy has on a young person's performance at school, up to the age of 15 when they get their GCSE results (and results from other equivalent qualifications); we will try to find out if this truancy continues to have an impact even when they leave school. Consider another example: we will shed new light on the extent to which disadvantaged young people, with a good set of educational choices facing them at age 15, are seen to make 'bad' choices; when compared to their more advantaged peers, facing the same choice sets. Similarly, the study will shed light on the choices made by young people from age 16+ who are from more advantaged backgrounds, who we see facing a more limited set of educational choices at 15; and how these differ to young people from disadvantaged backgrounds facing the same limited choices.

Planned Impact

Pathways to Policy Impact: As the current study develops, Prof. Urwin and Prof. Gregg will use their continuous engagement with policy colleagues across BIS, DWP and DfE to ensure the work described in this submission remains policy relevant; and that the findings are disseminated widely across policy communities. More specifically, over the lifetime of the project, the accompanying letters of support suggest access to relevant policy colleagues, who we would engage with, to feed into the research questions being tackled. We would expect to begin production of outputs around five months into the project, following initial matching of NPD-ILR-HESA data. We will receive support in the form of organisation and hosting of cross-departmental events, as outputs are developed. During the initial 5 months of the project we will discuss the work with Resolution Foundation colleagues; and then design events and outputs that would be supported by the foundation over the remaining months and beyond.

Pathways to Practice Impact: The work described in this submission will benefit from Prof. Urwin's continual engagement with practitioner bodies, such as the 157 Group, Federation of Awarding Bodies, CIPD and Holex. In addition to these bodies, we would additionally attempt to engage with colleagues from the Association of Colleges, the Association of Employment and Learning Providers, the Learning and Work Institute, the Education and Training Foundation, the Headteachers' Roundtable, the Career Development Institute, OFSTED and the main teaching unions - possibly in the first instance, via invitations to an event hosted by the Resolution Foundation.

Prof Urwin is currently advising Skills Development Scotland on how Scottish educational datasets might be used to carry out the sort of analysis we have carried out for England; and Dave Thomson has worked on production of post-16 value added indicators for schools and FE in Wales. We will use these links to ensure that our research findings have impact beyond the geographical focus of data analysis, which will be NPD cohorts of young people in England. In addition, Prof. Urwin and Prof. Gregg have presented at OECD Global Forums in recent years; colleagues at Westminster are currently working on parts of the European Qualifications Framework; and Prof Gregg has written on issues of unemployment for the OECD and EU. As the work develops we will engage with our various contacts in these international bodies to ensure that the work has impact beyond the UK.

A key aim of the research is identification of impacts from pupil background characteristics on performance, but also choices made through the post-compulsory education system. As well as alerting policy and practice communities to the results of this research, we will communicate messages to young people and their families, via a range of media.

Administrative data communities: DfE, in partnership with DWP and BIS, are currently undertaking a process of data matching across the datasets that form the focus of this study; and our programme of work takes this into account. Datalab currently host the NPD User Group and Dave Thomson is the UK's recognised NPD expert; Thomson developed the ILR-WPLS matched database currently used by BIS [developing the majority of underpinning scripts]; and also the ILR-WPLS-NPD-LMS data held by DWP. As the work on data matching progresses, we will engage with a variety of data users and the team would aim to create an ILR User Group, that would consider issues across these various datasets (also feeding into DWP's WPLS user group).

Publications

10 25 50
 
Description We have created versions of the LEO [Longitudinal Educational Outcomes] administrative data to investigate Young People at Risk of Poor Educational and Labour Market Outcomes; across cohorts who turn 16 in the academic years from 2001/2002 to 2008/2009. The LEO data link information on these cohorts from the National Pupil Database (NPD); the Higher Education Statistics Agency (HESA); the Individualised Learner Record (ILR); HMRC employment data (P45 and P14) and DWP Benefits information. The analyses focus on the half of young people who are unlikely to pursue university education and considers issues of social mobility:

1. Creation of datasets for analysis is a key achievement, as young people who do not take a standard 'academic' (GCSE - A-level - Degree) pathway, choose between a greater variety of institutions (General FE, School Sixth Form, College Sixth Form) and qualification routes (including Apprenticeships and, during the period under study, NVQs). The datasets created provide structure for analysis, retaining salient aspects of the post-16 learning environment, whilst allowing systematic analysis. We are engaging with DfE (the data owners) to share lessons from data development and analysis.
2. One version of this LEO data allow investigation of disadvantaged young people with poor Key Stage 4 [KS4] performance - for instance, amongst the cohort who turn 16 in 2008/2009, 72% of pupils on Free School Meals [FSM] achieve a grade D or below in English and/or Maths GCSE. Analysis of subsequent pathways uncovers strong social segregation, with young people from poorer backgrounds predominantly following vocational/technical education pathways (where funding per pupil tends to be lower). The creation of institutional choice sets [from destinations of previous cohorts] and distance indicators that act as instrumental variables, allow us to estimate causal impacts on later life outcomes, arising from the post-16 educational choices of disadvantaged young people.
3. Working with the Low Pay Commission (LPC), we have carried out an initial evaluation of the impact of local labour market conditions on the employment and education decisions of young people; and the introduction in April 2016 of the National Living Wage (NLW) for those aged 25 and over, and the 21 to 24-year-old rate. Initial findings do not suggest local earnings and/or unemployment rates are economically significant in influencing young people's decisions, even when focusing on those particularly 'at risk' of being on minimum wages. There is some indication that the NLW may have boosted the employment prospects of young people well below the age 25 cut-off, but this is based on incomplete LEO data, that we are attempting to supplement via an ongoing data request to DfE.
4. A number of blogs have focused on specific aspects of the education system. For instance, Long-term outcomes: How did life turn out for pupils who took GNVQs? and a similar analysis considering those, who took Level 3 BTEC qualifications. We have presented findings on the link between exclusions, alternative provision and off-rolling and initial investigation of the educational impacts arising from a Raising of the Participation Age point to insignificant results.
Exploitation Route We have met with members of DfE Skills Analysis team (and DWP colleagues) and have been invited to present our report to DfE on pathways and impacts for young people who form the focus of analysis under [2.]. This is the first stage in a process of engagement, which will also consider data issues [under 1.]. There is currently no LEO User Group and such groups have been important in developing standardised approaches to development of admin data analysis.

Under 3. we are currently requesting additional data from DfE to supplement the analysis already presented as part of the Low Pay Commission Research Symposium in September. The final analysis will be part of the evidence base informing the LPC Review of Youth Rates in Spring 2019.

As these policy-focused outputs are developed, specific components will be written up for submission to peer reviewed journals and academic conferences - for instance, the analysis estimating causal impacts on later-life outcomes arising from post-16 institutional choice. In addition to this engagement with policymakers and academics, we will continue the wider programme of public dissemination already begun with blogs and podcasts (detailed elsewhere); with a dissemination event planned for summer 2019.
Sectors Education

URL https://ffteducationdatalab.org.uk
 
Description An assessment of what factors affect young people's decisions to enter the labour market
Amount £30,000 (GBP)
Funding ID CR18059 
Organisation Low Pay Commission 
Sector Private
Country Unknown
Start 05/2018 
End 05/2019
 
Description "Does the Minimum Wage Impact Labour Market Choices of Young People aged 16 to 24? Emerging Findings", Low Pay Commission Research Symposium, 6th September 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Policymakers/politicians
Results and Impact Presentation of emerging findings from analysis of LEO to inform deliberations of LPC Commissioners
Year(s) Of Engagement Activity 2018
 
Description Blog Datalab website: "Long-term outcomes: How did life turn out for pupils who took Level 3 BTEC qualifications?" 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Policymakers/politicians
Results and Impact The blog on Datalab website uses LEO data: Over the last ten to fifteen years, there has been a substantial increase in the number of young people achieving level 3 (A-level and equivalent) qualifications. Much of this growth, particularly so for disadvantaged students, has been due to level 3 BTEC qualifications. Yet despite their increased prevalence, their value as currency for higher education admission has been called into doubt. Strictly speaking, BTECs are considered to be "applied general" qualifications. Perhaps their status as a middle ground between academic qualifications and vocational qualifications is part of the problem. So if you are a young person who has just completed Key Stage 4, and you are considering higher education, would you be better off taking A-levels or BTECs? No impacts yet known.
Year(s) Of Engagement Activity 2018
URL https://ffteducationdatalab.org.uk/2018/11/long-term-outcomes-how-did-life-turn-out-for-those-who-to...
 
Description Datalab Blog: "Long-term outcomes: How did life turn out for pupils who took GNVQs?" 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Policymakers/politicians
Results and Impact GNVQs were introduced in the early 1990s to provide experiences and education relevant to working life, but stopped short of providing training for a specific job. They could be studied at foundation level (equivalent to GCSE grades D-G), intermediate level (equivalent to GCSE grades A*-C) and advanced level (equivalent to A-Level), and were available in 15 broad vocational areas. The full GNVQ consisted of six modules, each of which had a recommended guided learning time of 60 hours, or else a qualification consisting of half this number of modules (a 'part one GNVQ') could be done. Two-thirds of the modules were internally assessed based on coursework. The full GNVQ was considered equivalent to four GCSEs and the part one GNVQ equivalent to two GCSEs. By 2003, pupils at the end of Key Stage 4 were entering some 135,000 foundation and intermediate GNVQs. Over 60% were in applied information and communication technology. This blog uses LEO data to describe the relative long term outcomes for this group of learners
Year(s) Of Engagement Activity 2018
URL https://ffteducationdatalab.org.uk/2018/10/how-did-life-turn-out-for-pupils-who-took-gnvqs/
 
Description FFT Datalab Blog Post: "Long term outcomes: Does school quality matter?" 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Policymakers/politicians
Results and Impact Now that data from the National Pupil Database has been linked to earnings, employment and benefits records [as part of LEO] we can begin to examine some really big questions about the long-term effects of education policy, structures and qualification choices. In this post, we look at whether school quality makes a difference to pupils' long-term outcomes.
Year(s) Of Engagement Activity 2019
URL https://ffteducationdatalab.org.uk/2018/12/long-term-outcomes-does-school-quality-matter/
 
Description FFT Datalab Blog Post: "Long-term outcomes: Does it make a difference if your school has a sixth form?" 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Policymakers/politicians
Results and Impact In this post, we look at whether pupils who attend a school with a sixth form achieved better long-term outcomes than pupils who went to schools without a sixth form.
Year(s) Of Engagement Activity 2019
URL https://ffteducationdatalab.org.uk/2018/12/long-term-outcomes-does-it-make-a-difference-if-your-scho...
 
Description FFT Datalab blog post: "The link between exclusions, alternative provision and off-rolling" 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Policymakers/politicians
Results and Impact Over the years FFT Datalab have looked at exclusions, alternative provision and, in particular, pupils disappearing off school rolls. In this blogpost, the team use LEO data to examine the intersection between these three things.
Year(s) Of Engagement Activity 2018
URL https://ffteducationdatalab.org.uk/2018/11/the-links-between-exclusions-alternative-provision-and-of...
 
Description SHARE Radio podcast: Policy Matters meets Economist Questions, Social Mobility and Further Education 
Form Of Engagement Activity A broadcast e.g. TV/radio/film/podcast (other than news/press)
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Public/other audiences
Results and Impact This episode of Policy Matters is a cross-over show in which hosts Franz Buscha and Matt Dickson are joined by the host of Economist Questions, Peter Urwin. As Peter is currently leading a large research project looking at young people's pathways through education, Franz and Matt ask him about his own journey and how that affected his social mobility. They go on to discuss the problems that the Further Education system faces in providing both second chances for those who don't achieve well at age 16 as well as higher-level training for those more suited to the vocational route. All this in the context of dwindling education budgets in general, and a lack of policy focus on the Further Education system.
Year(s) Of Engagement Activity 2018
URL https://www.shareradio.co.uk/podcasts/?pr=Policy+Matters