1. Evidence-informed educational practice for children in care
Dr Alun Rees & Lucy Wawrzyniak
Visiting Practitioner Fellows
Rees Centre for Research in Fostering and Education
University of Oxford, Department of Education

2. Aims for the webinar Outline recent research findings
Put the findings in a practitioner context
Schools
Virtual schools
Other children’s workforce professionals
Challenge you to think about your place in developing the evidence base

3. Why bother - SCHOOLS
Looked after children are a specific priority for Ofsted
School Inspection Handbook: paragraphs … 59
129
171
195
210
59. When considering performance data, inspectors should take account of specific groups such as looked after children …
129. It is important to test the school’s response to individual needs by observing how well it helps all pupils to make progress and fulfil their potential. It may be relevant to pay particular attention to the achievement of … disadvantaged pupils, including looked after children.
171. Inspectors must look at a small sample of case studies in order to evaluate the experience of particular individuals and groups, such as … looked after children and those with mental health needs.
195. Inspectors must take account of … pupils’ progress in the last three years … including that of looked after children
210. … inspecting and reporting on students’ achievement in the sixth form, inspectors must take into account all other guidance on judging the achievement, behaviour and development of students, including specific groups such as … those who are disadvantaged (In the sixth form, the term ‘disadvantaged’ applies to looked after children and students who were known to be eligible for free school meals when in Year 11)

4. Why bother – LOCAL AUTHORITY
‘Promoting the education of looked after children - statutory guidance for local authorities’, July 2014
Local Authority Inspection Framework: what is ‘Good’ …
… attend school or other educational provision and they learn.
… assessments of their needs, as well as specialist support … help them to make good progress in their learning and development …
… receive the same support from their carers as they would from a good parent.
… attainment gap between them and their peers is narrowing.
… up-to-date information about how looked after children are progressing at school … urgent and individual action when not achieving well.
Social workers take note – you are not going to get good judgements on the basis of care and safeguarding alone. Educational outcomes and practices are a specific risk to good judgements. VSH are being quizzed just like bricks and mortar school heads and inspectors are increasingly looking at data showing how care decisions are influencing education outcomes.

5. Why bother – VIRTUAL SCHOOL
I’ve left 2014 out at this point because I’m not certain I understand the impact of the significant changes in the goal posts and they certainly make meaningful comparison with 2013 tricky
Outcomes, and trends in outcomes, are too varied.

6. Research … what can you believe?
Best buy
Randomly Controlled Trial (RCT)
Comparison with robustly constructed control group
Correlation & Statistically Significant associations
Association vs. Impact
Worth looking at
Validated local practice
Promising local practice
Emerging local practice
Don’t buy
Blog posts and twitter feeds
… use them as prompts

7. What are the risk and protective variables for the education of children in care? Aoife O’Higgins
Purpose: To inform the resource priorities of central and local government, the practice of professionals, and the databases used to monitor outcomes.
October 2013 – October 2016

8. Systematic Review – Aoife O’Higgins
37 quantitative studies from US, England & Canada:
Older children do less well in every study looking at age, whatever the age range studied, e.g. 9 year olds do less well than 7 year-olds as well as 16 year olds doing less well than 12 year olds.
Minority ethnicity associated with poorer outcomes.
Boys do worse, on average.
Children with SEN and behavioural issues do less well (and the proportion is much higher in the looked after cohort).
Findings on age at entry into care, stability and placement type are mixed, but the different study designs place a question mark over these findings.
Carer attitudes, aspirations and home-based involvement consistently predicted better educational outcomes
In many ways the looked after cohort reflect the whole cohort – the factors influencing looked after children are, largely the same as those influencing their peers – though more so.
While experience on the ground suggests age of entry and stability and placement type are important, the existing research base isn’t convinced – finding are mixed, but that might be as much to do with research designs as reality.
Where the research is consistent is in relation to carer attitudes, aspirations and involvement: they consistently preduct better outcomes.
DfE data is more convinced about time in care and placement turbulence …
NB: preliminary exploratory analysis

9. The Educational Progress of Looked After Children in England: Linking Care and Educational Data
Purpose: To inform the resource priorities of central and local government, the practice of professionals and the databases used to monitor outcomes.
February 2014 – April 2015
School for Policy Studies

10. Research design
Linking national data sets on the education (National Pupil Database) and care experiences of looked after children in England (SSDA903)
to explore the relationship between educational outcomes, the children’s care histories and individual characteristics, and practice and policy in different local authorities.
Interviews with 36 children in six local authorities and with their carers, teachers, social workers and Virtual School staff
to complement and expand on the statistical analyses, and to explore factors not recorded in the databases (e.g. foster carers’ attitudes to education, role of the Virtual School)

11. Description of the cohort
7,852 looked after children eligible for GCSEs in 2013
4,847 had been in care for 12 months or more continuously, of which:
44.2% female
78.8% White British
51.3% first entered care aged 10 or over
Looked after children are over-represented in indicators of difficulty:
CLA for 12+ months
Non-care
SEN Statement
30.3
3.2
% of SEN with BESD
50.4
26.6
Attending PRU
12.3
1.1

12. How might these difficulties impact cohort performance?
Ave. Key Stage 4 points*
SEN Statement
199.1
No special need
368.0
Identified BESD
247.5
356.8
Attending PRU
105.3
Average across all school types
342.4
*Key Stage 4 points: 8 best GCSE, or equivalent; range of 0-464)
Therefore, CLA cohort statistics are significantly driven by identified special needs and their resulting school placements

13. Proportion of the cohort
‘Care Career’ types
Career Type
Proportion of the cohort
Entry aged 0-4
14.8
Entry aged 5-9
30.2
Entry as adolescent (‘other reasons’)
21.3
Entry as adolescent (‘abuse or neglect’)
24.0
Entry as unaccompanied asylum seeker
3.4
Entry as ‘disabled’
6.4
Not in care
(for comparison)

14. Proportion of the cohort Average Key Stage 4 Points
‘Care Career’ types
Career Type
Proportion of the cohort
Average Key Stage 4 Points
Entry aged 0-4
14.8
217.7
Entry aged 5-9
30.2
229.0
Entry as adolescent (‘other reasons’)
21.3
185.5
Entry as adolescent (‘abuse or neglect’)
24.0
211.4
Entry as unaccompanied asylum seeker
3.4
232.7
Entry as ‘disabled’
6.4
47.7
Not in care
(for comparison)
343.5
Other reason include risk taking, criminal activity drug abuse etc.

15. Placement history
For the 4,847 who had been in care for at least the last 12 months:
16.4% had been in care for under 2 years; 25.6% for 9 years or more;
29.0% had been in most recent placement for under a year
36.4% had always been in foster care; 3.0% always in kinship care; 3.5% always in residential care;
Fewer KS4 placements were foster care than at KS2 (59.6% vs. 70.2%), use of residential care increases (18.5% vs %)
17.3% had only had one placement; 10.2% had had 10 or more placements

16. Number of placements & KS4 points
BUT we are analysing this alongside other variables to get the full picture of what really counts

17. Type of school placement
Care ‘Career’
Type of school placement
Number of placements
We have from the analysis to date is a more sophisticated understanding of the associations between these factors and key stage 4 points score.
The next step is to begin to look at the ways in which these factors interact to influence points score. For instance might it be that PRUs aren’t as disastrous if the child is in a secure foster placement, or could certain ‘careers’ be more or less at risk from some of the other factors at play?

18. Questions & Comments?

19. What (might) work … Education Endowment Foundation Toolkit
Considering the needs of Looked After Children:
Aspiration Interventions – how do we support carers to bridge the gap between aspiration and achievement
Supported work at home
Mentoring
Encourage you to type in your questions as you go so we can pick them up at the end.
Now I’ll hand over to Lucy for the second half

20. What (might) work … Aspiration
Shown to have low impact for high cost in the EEF toolkit, most likely because most children don’t have low aspirations.
Cheung et al. (2012) and Flynn at al. (2013) found a correlations between the aspiration of the carers and attainment.
How do we support carers in their aspirations for their young people and how do we as all professionals support young people in achieving those aspirations?

21. What (might) work … Support at home
Not found to be particularly effective in primary children by the EEF but …
Caregiver support at home had a positive impact on outcomes (Burley and Halpern 2001; Cheung et al., 2012 and Pears et al., 2010)
Can have particularly positive impact on girls in transition when carers are involved. (Pears et al. 2012).
How can carers be supported by schools to support work at home?

22. What (might) work … Mentoring
EEF: low impact for moderate cost – might be dismissed.
‘Handbook for Youth Mentoring’ (O’Hara), suggests there may not necessarily be academic gains but the ability of the mentor to build relationships can be highly beneficial if they persevere.
If the mentor is prepared to have their commitment challenged and understands that if they miss a session they are reinforcing rejection, the potential gain is great in terms of the young person experiencing another positive relationship.
The barrier can be the young person could feel they have more to lose than gain in the outset.

23. What (might) work … One-to-One tuition
Berridge (2009) both self-esteem and application to study improved.
Other studies showing positive impact ( Flynn et al 2012, Liabo, Gray and Mulchay 2012 and Forsman & Vinnerljung 2012).
The same challenges as mentoring are likely to apply (consistency, relationships etc.) (Britner et al.,2014)

24. What (might) work … Collaborative learning
Comes out strongly in the EEF and backed up in relation to LAC through Slavin (2010).
Not always ‘what’, but ‘how’

25. So …
Avoid placement disruption – quality of school and carer understanding/relationship?
Avoid Key Stage 4 moves - is the change REALLY necessary?
Seek a consistent plan that everyone, including the young person, is signed up to
How can the team around the child support these?

26. Could you do things differently?
How could you structure carer involvement
Tuition outside the school day in the carer’s home
Paired reading supported at home
Joint carer, social worker and school training
Hampshire – direct effect on reading in primary phase
Bob Flynn – indirect effect on emotional health

27. Questions & Comments
Carer attitudes, aspirations and home-based involvement consistently predicted better educational outcomes
Carer involvement can have particularly positive impact on girls in transition when carers are involved
Working WITH … delivers effective, lasting change
The strategies with the strongest evidence base are tutoring (by carers or others), mentoring and supporting carers to support education – quality of ordinary care matters - the PPP enables all three to be done.
Listen to young people in care – not stereotyping, support and not identifying them as in care
What we know (or think we know) influences our behaviours and they impact on children’s outcomes
Improve social workers’ understanding of the factors driving educational outcomes and how their decisions will impact on them;
Improve teachers’ understanding of the factors that impact a child’s outcomes and the influence they can have on them
Improve carers’ understanding of the pivotal role they have in creating the stable home environment necessary for a child to learn
Bring social workers, carers and teachers together to increase mutual understanding

28. How you can be involved
Express interest in being involved in future possible research projects;
Come along to lectures, seminars/webinars;
Join the Rees Centre mailing list and receive newsletters 5 times/year:
Web:
Comment on the Centre blog – or write for us;
Follow us on

29. Challenging our own thinking – Knowledge Claims
Reading at home with foster carers has a positive impact on outcomes
Reducing exclusions increases the educational outcomes for looked after children
A joined up consistent approach is essential in improving the educational outcomes for looked after children
‘Ordinary care’ provides an the basis for successful interventions
Children and young people with input into their care plan will achieve better outcomes

30. How can you contribute? The Virtual School Handbook
Identify the best practice and use it
The DIY Evaluation Toolkit
Bring rigour to decisions about what to do and the evaluation of how it worked
Professional networks
bring designated teachers together with social workers and carers and establish a joint learning enterprise; talk to your teaching school alliance

31. Acknowledgements ESRC and University of Oxford for the grant;
Judy Sebba for including us in it;
Nikki Luke and Aoife O’Higgins for their generosity of time and effort
Andrea Diss and Sally Winiarski for their practical support