Sebba o higgins-educational outcomes of children in care_4_nov2014

Understanding the Educational
Outcomes of Young People in Care
Judy Sebba and Aoife O’Higgins
The Rees Centre

What the session will cover
• The Rees Centre
• The state of research in the children’s social
care field
• Some evidence on what influences
educational outcomes in young people in care
• Some emerging preliminary findings from the
Nuffield-funded research on educational
progress

Rees Centre for Research in Fostering and
Education
The Rees Centre aims to:
• identify what works to improve the outcomes
and life chances of children and young people
in foster care
We are doing this by:
• reviewing existing research in order to make
better use of current evidence
• conducting new research to address gaps
• working with service users to identify research
priorities and translate research messages into
practice
• employing foster carers and care experienced
young people as co-researchers
Centre is funded by the Core Assets Group but
has grants from a range of other funders

The current evidence base in children’s services
(adapted from Stevens et al, 2009, p.286)
Methods (selective)
No of
used in 625 studies
studies
% of
studies
Qualitative 230 37
Mixed method 108 17
Longitudinal 74 12
Quantitative dataset analysis 16 3
Non-randomised trial 8 1
RCT 3 <1
Systematic review 2 <1

Analysis of designs used in studies
• Analysis based on 5 research reviews in Rees Centre
series, a much larger review on mental health
interventions in looked after children for NSPCC and
Aoife O’Higgins systematic review;
• Our reviews are essentially ‘what works’ reviews so
published reviews are referred to in the conceptual
background and discussion but not included in the
in-depth review;
• Unlike some well-established systems for systematic
reviewing we do not exclude studies on the basis of
their methodological approach but we do attempt to
give more robust studies greater weight in the
conclusions.

Studies included in Rees Centre reviews (categories not mutually exclusive)

Review No. of
studies
Qualitati
ve
%
Survey
(n >30) %
Longitudin
al
%
Sec data
analysis
%
RCT
%
Non-random

trial %
Pre and
post
meas %
Mixed
method
%
Stevens et al
(2009)1
625 37 4 12 3 4 1 1 17
Motivation to
foster
32 31 59 3 (1) 6 (2) 0 0 3 (1) 9 (3)
Peer support 33 88 36 0 3 (1) 0 0 0 15
Impact carers’
children
17 71 47 0 0 0 0 0 18 (3)
Parent & child
fostering
27 81 11 (3) 0 18 0 0 0 22
Role of the
SSW
22 81 45 0 9 (2) 0 0 0 32

Mental health
interventions2
75 12 1 (1) 0 0 52 15 19 9
Factors
associated
educ.outcome
34 0 44 6 (2) 50 0 0 0 0

Issues and implications
• The evidence base does not yet allow us ‘to intervene
with the most benefit and the least harm’
• A key criterion in making decisions about research
designs must be ‘fitness for purpose’ but the basis for
national policy making (e.g. RCTs & replications) and the
need by practitioners for quick answers that tell them
what to do can be in direct conflict;
• The questions we address in the Rees reviews are all
generated from the field which limits the number of RCTs
used in the studies that address them;
• Possible that large scale surveys are being replaced by
secondary data analysis as the development of large
scale datasets and access to them improve.

• So we want to describe some work linking
large datasets in order to identify what
influences educational outcomes.
• But before we do this, let’s get a perspective
from a young person who has experienced the
care system.

The schooling experience of a young person in
care
“I entered care at the age of 5. By the time I was 6 and had
begun primary school I had had 15 different foster placements.
That’s 15 different houses, sets of rules and family values. Early
school life was very difficult from being left at school late until
someone collected me to turning up late without the correct
equipment or kit for that day because I had spent the night upset
in a home I didn't know.”
“Secondary school was no easier the only difference being I was
no longer interested in education but instead my only goal was
to go home to my mum. I’ll never forget the receptionist at my
school as she was the only consistent person in my life and
always remembered to ask how my mum was doing.”

Educational outcomes of looked after children in England
DfE (2013b) Statistical First Release 11 Dec 2013
• 15% achieve expected grades at 16 years (Key Stage 4) compared to
58% of all children – a 43% gap
• Twice as likely to be permanently excluded
• Three times as likely to have a fixed term exclusion
• BUT achievement gap is smaller at Key Stage 2 (age 11) (26% for
Maths, 23% reading, 28% writing)
In addition
• Only 7% access Higher Education (DfE, 2013a) compared to around
50% of general population
• Educational experiences and outcomes contribute to later health,
employment (22% unemployment rate), involvement in crime
(27% of those in prison; Social Exclusion Unit, 2002)

Educational outcomes of looked after children in England
(Source: DfE, 2013)
45
40
35
30
25
20
15
10
5
0
12 to 18
months
18
months to
2 years
2 to 3
years
3 to 4
years
4 to 5
years
5 to 6
years
6 years or
more
Percentage achieving
Length of time in care
5+ GCSEs at grades A*-C
5+ GCSEs at grades A*-c
including English and
mathematics

Research on the education of children in care
International research confirms the attainment gap between
children in care and their peers (Trout et al., 2008) and the high
prevalence of special educational needs and exclusions (Scherr,
2007).
The reasons for this are not clear:
- Is the care system detrimental to children’s education?
- Are there pre-care factors which predispose children to low
educational outcomes?

Pre-care factors
• 75% from single parent families
• 75% receive benefits
• 80% are in rented accommodation
• 60% living in poor neighbourhoods
(Bebbington and Miles, 1989)
Maltreatment: abuse, neglect, abandoned, behavioural
problems… (Eckenrode, 1993; Shonk & Cicchetti, 2001)
What is the impact of these events on educational outcomes of
children in the general population?

What happens in care?
• Mental health problems (Perzow et al., 2013)
• Behavioural problems (Colton & Heath, 1992; Flynn et al.,
2013)
• Learning impairment (Geenen & Powers, 2006)
• Placement instability (Wise et al., 2010; Zima et al. 2000)
• School changes (Burley & Halpern, 2001)
• Low expectations (Heath et al., 1993)
• Risk taking behaviour, e.g. substance misuse. (Shin, 2003)
• Caregiver support (Pears et al., 2012)
• Caregiver expectations and aspirations (Flynn et al., 2013;
Wise et al., 2010)
• Caregiver home / school based involvement (Cheung et al.,
2012)

Some interventions to support the education of
children in care
In the UK:
• Letterbox Club (Griffiths, 2009)
• Paired Reading (Osborne et al., 2010)
Internationally
• Tutoring (Forsman & Vinnerljung, 2012)

The Educational Progress of Looked After
Children in Secondary Schools
Aim
To identify key care and educational factors that are
associated with the progress of children in care from
KS2 (end of primary school) to the end of KS4 (end of
secondary school) and their attainment at KS4
Purpose
To inform the resource priorities of central and local
government, the practice of professionals and the
databases used to monitor outcomes
In collaboration with University of Bristol and funded by Nuffield Foundation

Research design
How are we doing this?
• 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)

Dealing with the data
• Databases are constantly evolving
– outdated codes
– idiosyncrasies in data submissions
• Formatting data
– provided with a mix of episode vs. annual vs. individual level data
– ‘missing’ data (e.g. for Strengths and Difficulties Questionnaire)
– skewed data (e.g. children on a series of short-term respite breaks)
– creating variables from raw data (e.g. what does ‘placement length’
mean?)
– creating categories within variables (e.g. placements since KS2)

Database analyses
• Determined by research questions, theoretical relationships
and available data
• Regressions – which factors predict better or worse
educational outcomes?
• Multilevel modelling – what is the relative contribution of
factors at different levels?
Local authority
School
Individual child

Approaching the analysis
• Variable creation, analysis, interpretation
– an iterative process
• Describing the cohort
• Identifying key predictors
• Examining influences at different levels
– challenge of finding data
• Interpretation of findings
– looking at combinations of variables provides a clearer picture
– acknowledging the bias in our sample

Description of our cohort [preliminary analysis]
• 7,852 looked after children eligible for GCSEs in 2013
• 45.6% female
• 78.5% white
• 64.8% first entered care aged 10 or over
Career type Per cent
Entry aged 0 to 4 9.8
Entry aged 5 to 9 20.3
Adolescent entrant 29.5
Adolescent abused 26.5
Unaccompanied Asylum Seeking
Child
2.5
Disabled 11.4

Placement histories
• 26.7% had been in care for under a year; 16.8% for 9 years or more
• 25.3% had only had one placement; 4.0% had 11-15 placements
• 49.7% had been in most recent placement for under a year
• 36.5% had always been in foster care; 2.9% always in kin care;
15.0% always in residential care;
• Of those in care at end of primary school, 70.5% were in foster care;
11.2% kin care; 14.4% residential care
• Of those in care at end of GCSEs, 55.6% were in foster care; 7.0%
kin care; 32.1% residential care

Placement type
• Using ‘8 best GCSEs’ as an outcome (scores range from 0-464)
• Controlling for KS2 attainment, gender, ethnicity, age and reason for
first entry to care, and duration of last placement
• Compared to young people who left care during Year 11, young
people who remained in the following placement types did:
– foster care
• Better by an average of 41.0 points
– kinship care
• Better by an average of 11.8 points
– residential care
• Worse by an average of 29.3 points

Multi-level modelling
• Three-level model
– child; school; local authority
• Compared children in care for 12 months continuously with all
children
• Taking account of:
– KS2 attainment; gender; ethnicity; proportion eligible for free school
meals (school and local authority levels)
• Variation in KS4 attainment of looked after children at local
authority level was not statistically significant
– suggests that variability existed at the school level, rather than the
local authority level
• Two-level model revealed differences between local authorities

Qualitative Work
• Database analyses helped us identify six local authorities:
– three in which looked after pupils’ attainments and progress are
noticeably better than the average
– three in which they are worse
• Managers have been asked to identify six pupils = total
sample 36
– to avoid us matching them up to their NPD data
• We will ask young people if we can approach:
– foster carers/key workers/residential managers (36)
– social workers/family placement workers (36)
– designated teachers (36)
– Virtual School Heads (6)

Qualitative Work
• Interviews will be conducted by care leaver researchers and
foster carers
• indicative topics for interviews:
– identifying the type of support provided for looked after
children by schools
– engagement of foster carers/key workers/residential managers
and social workers with schools
– contribution of the Virtual School to improving outcomes
– backgrounds and attitudes of foster carers/key
workers/residential managers
– pupil motivation and behaviour (adults’ interviews only)
– involvement of Child and Adolescent Mental Health Services
(adults’ interviews only)

Informing policy and practice
• Project results will enable us to make recommendations, e.g.:
– timing of placement changes
– relative importance of placement and school stability
– relative importance of ‘quality’ and size in selecting schools
– selection and training of foster carers and residential staff
– use of data to predict performance and target interventions
– difficulties with current data system
– other useful forms of data that could be collected

References
• Conger, D., & Rebeck, A. (2001). How children’s foster care experiences affect
their education. New York City: Vera Institute for Justice.
• Department for Education. (2013a). Statistical First Release: SFR36/2013.
London: DfE. Retrieved from
http://www.gov.uk/government/uploads/system/uploads/attachment_data/fil
e/244872/SFR36_2013.pdf
• Department for Education. (2013b). Statistical First Release: SFR50/2013.
London: DfE. Retrieved
fromhttp://www.gov.uk/government/uploads/system/uploads/attachment_da
ta/file/264385/SFR50_2013_Text.pdf
• Flynn, R. J., Tessier, N. G., & Coulombe, D. (2013). Placement, protective and
risk factors in the educational success of young people in care: Cross-sectional
and longitudinal analyses. European Journal of Social Work, 16(1), 70–87.
• Pecora, P. J. (2012). Maximizing educational achievement of youth in foster
care and alumni: Factors associated with success. Children and Youth Services
Review, 34(6), 1121–1129.
• Social Exclusion Unit. (2002). Reducing offending by ex-prisoners. London: SEU.
• Stone, S. (2007). Child maltreatment, out-of-home placement and academic
vulnerability: A fifteen-year review of evidence and future directions. Children
and Youth Services Review, 29(2), 139-161.

Sebba o higgins-educational outcomes of children in care_4_nov2014

More Related Content

What's hot (20)

Similar to Sebba o higgins-educational outcomes of children in care_4_nov2014 (20)

More from Young Lives Oxford (20)

Recently uploaded (20)

Sebba o higgins-educational outcomes of children in care_4_nov2014

Editor's Notes