raising the attainment of eal pupils · 2% fle, no association with achievement for fle but -12...
TRANSCRIPT
Raising the attainment of EAL pupils
Briefing for potential applicants
Overview of the session
• Sir Kevan Collins, EEF: Introductions and overview
• Professor Steve Strand, Oxford University - EAL and
educational achievement in England: An analysis of the
NPD
• Eleanor Stringer (EEF Grants Manager) and Diana
Sutton (Director, Bell Foundation): Outline of the
funding round’s criteria, process and timeline
• Professor Therezinha Nunes, Oxford University:
Applying to the EEF – A grantee’s perspective
• Questions and discussion
• 4pm: Close
Who we are: EEF
The Education Endowment Foundation is an independent grant-making charity dedicated to breaking the link between family income and educational achievement.
The EEF was set up in 2011 by The Sutton Trust, as
lead charity in partnership with Impetus Trust (now part
of Impetus–The Private Equity Foundation)...
… with a £125m founding grant from the Department for
Education
The EEF and Sutton Trust are, together, the
government-designated ‘What Works’ centre for
improving education outcomes for school-aged
children.
The EEF evidence cycle
Teaching and
Learning Toolkit
EEF project
evaluations
Grant-funding Evaluation
Report
results
Synthesise
evidence
Who we are: Bell Foundation
• Cambridge based educational Foundation
• Started in 2012.
• Mission “Overcoming exclusion through language
education”
• Two thematic programme areas – children with English as
an additional language and offenders – Language for
Change programme
• Working in partnership with schools, charities and
universities through programmes and grants.
• Partner with EEF and Unbound on this funding round
• Follow us on twitter @bellfoundation
• www.bell-foundation.org.uk
Who we are: Unbound
Philanthropy
• Welcoming Newcomers. Strengthening Communities.
• Three priority areas: Legal Rights and Protection,
Integration, and Public Understanding.
• EAL is included under Integration.
• Endowed private foundation.
• Based in the US, with small office and staff in the UK.
• 51 active grants with a budget circa £2 million (UK).
Professor Steve Strand
9
EAL and educational achievement in
England: An analysis of the NPD
EEF briefing on Raising the attainment of
disadvantaged EAL pupils Funding Round
14 July 2015
Professor Steve Strand
University of Oxford, Department of Education
01865 611071
10
Research reports
https://educationendowmentfoundation.org.uk/uploads/pdf/EAL_and_educational_achievement2.pdf
11
Overview of this presentation
• Brief overview of the main points from the
summary report:
– Demographics EAL (size & distribution)
– Achievement profile by age 5-16
– Focus on Key Stage 2
– Risk factors for low attainment of EAL pupils
– Implications for policy & practice
12
Number of EAL pupils 1997-2013
• From 7.6% in 1997 to 16.2% in 2013
0
100,000
200,000
300,000
400,000
500,000
600,000
700,000
800,000
900,000
1,000,000
1,100,000
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
Nu
mb
er
of
pu
pils
Primary school age Secondary school age
13
EAL demographics
• Wide regional variation
(6% in SW and NE, 20%
West Midlands up to 55%
in inner London)
• Wide LA variation (17 of
top 20 in London but also
Slough, Luton &
Leicester, plus high in
Birmingham, Manchester)
• But also localised to
schools: of the 1,681
schools (8.4%) with
>50% EAL, 762 outside
London including NW,
Yorks & Humber ...
14
School Distribution %EAL • Very skewed: around one-quarter (22%) of schools
<1% EAL & over half (54%) <5% EAL
% EAL in school
N Schools
% Schools
0.0 - 1 4435 22.1 1.1 - 5 6346 31.7 5.1 - 10 2870 14.3 10.1 - 20 2240 11.2 20.1 - 30 1142 5.7 30.1 - 40 730 3.6 40.1 - 50 589 2.9
50.1+ 1681 8.4 Total 20033 100
• Widely dispersed: of the 1,681 schools with >50% EAL,
762 outside London including NW, Yorks & Humber
15
EAL gap by age - England 2013
(a) based on 241,545 students entered rather than whole cohort.
Source: DFE SFRs (2013) .
Age Stage Domain Measure
FLE
%
EAL
%
Odds
Ratio
5 EYFSP Reading At least expected level 73 63 0.63
Maths At least expected level 71 62 0.67Overall Good level of Development (GLD) 54 44 0.67
7 KS1 Reading Level 2A+ 57 48 0.70
Maths Level 2A+ 53 46 0.76
Overall Average Re + Ma (2A+) 55 47 0.73
11 KS2 Reading Level 4B+ 77 69 0.65
Maths Level 4B+ 74 72 0.90
GPS Level 4B+ 65.1 66.2 1.05
Overall Level 4B+ in RWM 64 59 0.81
16 KS4 English GCSE A*-C pass 68.8 64.6 0.83
Maths GCSE A*-C pass 71.2 71.8 1.03
Language GCSE A*-C (any language) 31.7 47.4 1.94
MFL(a) GCSE A*-C (French/German/Spanish) 67.4 74.9 1.44
Overall 5+A*-C Inc. En & Ma / Best 8 60.9 58.3 0.90
Overall EBacc achieved 22.5 24.4 1.11
16
EAL & Ethnicity • EAL closely correlated
with ethnicity (except for
Black Caribbean and
MWBC) but on its own
explains only 0.2%
variance in KS2 score
(ethnicity explains 1.8%)
• But EAL can add a little
(combined 2.2%) as
within each ethnic group
those with EAL
consistently lower
achieving (see next slide)
Ethnic group EAL % Total N
Bangladeshi 96.1% 9,410
Pakistani 88.4% 22,737
Any other ethnic group 86.0% 7,789
Any other Asian 81.5% 7,851
Indian 79.1% 13,437
Chinese 78.6% 1,758
White other groups 73.6% 22,579
Black African 70.9% 16,803
Traveller Gypsy/Roma 42.2% 1,451
Black other groups 36.6% 3,345
Other mixed background 23.5% 8,400
Mixed White & African 22.8% 2,703
Unclassified/Refused 17.9% 2,879
Mixed White & Asian 16.5% 5,080
Black Caribbean 4.2% 7,260
Mixed White & Caribbean 2.2% 7,130
Traveller Irish 1.6% 368
White Irish 1.6% 1,725
White British .4% 379,842
Total 17.3% 522,547
17
EAL and ethnicity – KS2 score 2013
-2
-1
0
1
2
3
Dif
fere
nce f
rom
KS
2 A
PS
po
pu
lati
on
mean
(28.6
) EAL
FLE
18
EAL and ethnicity
• White Other & Black African key because:
– Average EAL gap at KS2 is 2.5 NC mths, but for
Black African 5 NC mths & White Other 10 NC mths,
also large EAL gap for these groups at KS4;
– The two ethnic groups with the largest increase in
size between 2003 to 2013, both doubling in size
– Over 70% are EAL and over 40% recent arrivals in
UK (age 5-14) compared to 3% of all pupils (LSYPE)
• Looked at the top 10 languages (other than
English) within these ethnic groups and
compared achievement against the average for
FLE from the same ethnic group
19
White Other groups by First Language
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
Dif
fere
nce f
rom
KS
2 a
vera
ge p
oin
ts s
co
re (
AP
S)
for
Wh
ite
Oth
er
wit
h F
irst
Lan
gu
ag
e E
ng
lish
(F
LE
)
Note: Compared to White Other with FLE, adjusted for FSM, IDACI, migration & region (Full report p69-71).
20
Black African by First Language
Note: Compared to Black African with FLE, adjusted for FSM, IDACI, migration & region (Full report p69-71).
-4.0
-3.5
-3.0
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
Dif
fere
nce f
rom
KS
2 a
vera
ge p
oin
ts s
co
re (
AP
S)
for
Bla
ck
Afr
ican
wit
h F
LE
21
Other pupil risk factors
• Most risk factors broadly the same for EAL and
FLE - SEN, FSM, IDACI, age, gender
• But four have a particular risk for EAL:
– Ethnic group/First language: as described above
– International Arrival during the Key Stage (proxied
by absence of prior attainment score): 15% of EAL vs.
2% FLE, no association with achievement for FLE but
-12 NCmths for EAL
– Pupil mobility: e.g. Y6 EAL entrant -12 NCmths, FLE
Y6 entrant -4 NCmths
– Region: EAL in regions on average 4 NCmths lower
than London, but Yorkshire & Humberside 8 NCmths
22
School factors
-1.00
-0.75
-0.50
-0.25
0.00
0.25
0.50
0.75
1.00
Low (-1SD) Mean High (+1SD)
KS
2 A
PS
% students in school recorded as EAL
FLE
EAL
-1.0
-0.8
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
Low (-1SD) Mean High (+1SD)
CV
A s
co
re (
KS
2 A
PS
)
% students in school recorded as EAL
FLE
EAL
• Does a high %EAL impact negatively on achievement/progress
of FLE pupils? (Green, 2010; Cho, 2012)? Answer = No.
Low, mean & high %EAL represent -1SD, mean & +1SD in school distribution, specifically 0%, 15% and 33% EAL.
(Strand et al, 2015, p54/55)
23
Policy Implications • EAL in the NPD is a limited measure
“where a child was exposed to the language during early development
and continues to be exposed to this language in the home or in the
community. If a child was exposed to more than one language (which may
include English) during early development the language other than English
should be recorded, irrespective of the child's proficiency in English.”
DCSF (2006).
• EAL flag and fluency in English are not the same thing
– e.g. Strand & Demie (2005) report a direct measure of stage
of English fluency (Hilary Hester Scales) show that EAL pupils
who were fully fluent in English (over 40% of EAL pupils)
actually had higher achievement than monolingual English.
• Lack of fluency in English language is the real risk, this is
what is proxied by international arrival, school mobility
and ethnicity / first language
24
KS2 score & Stage of fluency in English
Inner London LA using Hilary Hester CLPE scales. 42% of EAL pupils were rated as fully fluent.
[Source: Strand & Demie, 2005].
25
Implications (Cont.) • Schools
– Need robust measures to identify fluency & need
– EAL Support materials: excellent resources were produced by
National Strategies e.g. Supporting pupils learning EAL (2002);
New Arrivals Excellence Programme (2007); PNS Learning &
teaching bilingual children in the primary years (2007)
• Funding
– Educational progress of EAL pupils is good, but cannot be
complacent as it reflects historically high levels of investment
(Section11 & EMAG until 2012/13)
– Post April 2013: 87% of LAs have a factor for EAL pupils in their
first three years at school (Inc. international arrivals): Min level
£466 primary & £1130 secondary.
– Broadly well targeted (though literature suggests may need 5-7
years for academic fluency)
26
References Cho, R. M. (2012). Are there peer effects associated with having English Language Learner (ELL)
classmates? Evidence from the Early Childhood Longitudinal Study Kindergarten Cohort (ECLS-K).
Economics of Education Review, 31(5), 629-643.
DCSF (2006). Pupil Language Data: Guidance for local authorities on schools' collection and recording of
data on pupils languages. London: DCSF.
Demie, F. & Strand, S. (2006). English language acquisition and educational attainment at the end of
secondary school. Educational Studies, 32, (2), 215-231. http://dx.doi.org/10.1080/03055690701423184
DFES. (2007). Learning and teaching for bilingual children in the primary years: Teaching units to support
guided sessions for writing in EAL (Ref 00068-2007). London: Department for Education and Skills.
NALDIC (2015). EAL Funding. http://www.naldic.org.uk/research-and-information/eal-funding
Geay, C., McNally, S., & Telhaj, S. (2012). Non-native speakers of English in the classroom: What are the
effects on pupil performance? Centre for the Economics of Education (CEE). London School of
Economics, London. http://www.nuffieldfoundation.org/sites/default/files/files/ceedp137.pdf
Strand, S. (2010). Do some schools narrow the gap? Differential school effectiveness by ethnicity, gender,
poverty and prior attainment. School Effectiveness and School Improvement, 21(3), 289-314.
http://dx.doi.org/10.1080/09243451003732651
Strand, S., & Demie, F. (2005). English language acquisition and educational attainment at the end of
primary school. Educational Studies, 31, (3), 275-291. http://dx.doi.org/10.1080/03055690500236613
Strand, S., Malmberg, L., & Hall, J. (2015). English as an additional language and educational achievement
in England: An analysis of the National Pupil Database. London: Educational Endowment Fund.
http://educationendowmentfoundation.org.uk/uploads/pdf/EAL_and_educational_achievement2.pdf
27
End of Presentation
Raising the attainment of
disadvantaged EAL pupils:
The criteria, process and timeline
Eleanor Stringer, EEF
Diana Sutton, Bell Foundation
Initial evidence review
• EEF, Unbound Philanthropy and The Bell Foundation worked
together to fund an initial analysis of the extent and nature of
the issue (Professor Strand’s work).
• This included a review by Professor Victoria Murphy, which
sought to:
– identify and review controlled intervention studies which
focused on and/or included EAL pupils’ English language
and literacy development
– identify the quality of these studies re: their contribution to
understanding teaching and learning for EAL pupils
– identify intervention programmes which might be most
suited to the UK context
Summary/conclusions of Professor Murphy’s research
• Very limited research outside US context
– Need for more controlled intervention studies particularly for the most at risk sub-groups.
• Most interventions included some form of explicit vocabulary/word-level instruction
– Teaching vocabulary, particularly Academic Vocabulary, can have positive effects on children’s reading and comprehension
• Children who are ‘at risk’ and/or struggle with reading accuracy can benefit from interventions targeting word analysis
– Strand’s analysis identifying more at risk sub-groups in the UK context might enable us to provide more targeted support
• Students who are good readers but are weak in comprehension can benefit from interventions which target explicit vocabulary teaching
• A comparative lack of interventions looking at Continuous Professional Development and in-service training for teachers
– Teachers need support in their work with children with EAL and we need to offer them (more) evidence-based in service training
• The role of the home and home language environment is under-represented in intervention research
– More research exploring how to effectively support home language knowledge in linguistically diverse classrooms in the UK
Summary/conclusions of Professor Murphy’s research
Overview of the call for proposals
• The three funders are putting £2m into a fund for trialling
interventions that:
1. Focus on improving on improving learning outcomes for
these pupils, and have some existing evidence of
positive impact on their attainment;
2. Could be funded to work across a number of schools in
England;
3. Could be evaluated by an independent evaluation team;
and
4. Have the potential to be scaled-up further if shown to be
cost-effective.
1: Focused on attainment and evidence-based
• We are looking for proposals that draw on the existing
evidence about “what works” for pupils for
underperforming or disadvantaged EAL pupils
• We will prioritise interventions with evidence that
suggests that pupil attainment increased by more than it
would have done otherwise, i.e. through comparison
with a similar control group.
• At the very least, we expect interventions to informed by
the wider evidence base.
2: Could be funded across a number of schools in
England
• In order to assess the impact of an intervention, we often
need to involve a large number of schools (usually at
least 15, sometimes many more)
• We need to be clear what the intervention is – if it’s too
vague, then it will be hard to implement in other schools.
• We are not a source of funding for continuation of a
service in a single school, or small group of schools.
• Grantees will need to have the skills to recruit, manage
and deliver to a range of schools.
3. Independent, rigorous evaluation
All projects evaluated by a member of our 26-strong panel of evaluation experts (including universities and research organisations). These will be selected once the most promising proposals have been identified. We expect grantees to collaborate with the evaluation teams and funders to develop the project and evaluation plan. This could involve changing the initial proposal substantially (e.g. streamlining the intervention; changing the year group; expanding the number of schools involved)
• Robust designs, mostly trials
The impact of most interventions will be evaluated through a “randomised controlled trial” – half the schools/pupils participate, the others are a control group to compare outcomes.
• Quantitative measures
Effect on attainment (e.g. GCSE results, literacy tests) and cost—so we can compare and contrast between projects. The exact measure will be decided with the project teams. All reports will be published, no matter the results.
• Qualitative and process evaluations also crucial
To find out if/how it works in real world school conditions, evaluators will often interview/survey teachers
Examples of EEF reports
4. Have the potential to be scaled up further
• As well as being able to be evaluated in a number of
schools (criteria 2), we want to find interventions that – if
successful – could be delivered in schools across the
country.
• This means that they shouldn’t be dependent on a
particular local resource or delivery team, unless these
can be replicated elsewhere.
37
Process and Timeline
38
Deadline for applications 1st October 2015
Grants teams review applications against criteria
Contact applicants with initial questions
The first Grants Committee
(EEF, Bell and Unbound all involved)
December 2015
Further discussions with successful applicants
Appoint an independent evaluation team
Detailed meetings with evaluators
The second Grants Committee and
final approval from the Board of
Trustees of all funders
March 2016
Further discussions with evaluation team and funders
Draw up a grant agreement, payment linked to milestones (e.g. recruiting x
schools to the project
Some projects ready to begin trials in
schools Summer 2016
The most important questions
• We expect to receive a large number of applications (we typically receive 100+), but will only be able to fund a small number of promising projects.
• The most important parts of the application form are questions:
– 2.2: Outline the proposed project in simple language, clearly explaining what the intervention is (e.g. What type of teaching practice is use? How are the teachers trained? How often is it delivered?)
– 2.5: Outline the evidence: Focusing on quantitative attainment data, ideally showing impact compared with a control group
– 2.7: What do you think are possible ways that the proposed intervention could be taken to scale?
39
The rest of the application form
• We expect applicants to be open to input from the
funders and the evaluators to shape the project.
• While we ask for ideas of the number of schools you can
work with, and for the cost of the project, we will expect
these to change and use them to give us an initial
indication of scale and cost.
• Therefore do not worry too much about the detail in the
other questions of the form.
40
Questions to consider
• Is your intervention ready to be evaluated? Can it be
clearly defined and delivered?
• Is your intervention likely to show an impact on
attainment when subjected to a randomised controlled
trial?
• Does your team have the capacity to deal with the
demands of running RCT? E.g. delivering to many
schools at the same time.
41
Professor Therezinha Nunes
Oxford University
Improving Numeracy and Literacy in Key Stage 1
• Background about the intervention
– What it is about and previous research
• Application form and process
– say the crucial things about causal relations very clearly
– each sentence has to make a point
– think about design but be prepared to think again
• Process following the application
• Working with the evaluators
• What potential applicants might want to think about
– Learning from the EEF
Terezinha Nunes, Peter Bryant, Deborah Evans, Rossana Barros - Department of Education, University of Oxford
EEF evaluation report: Jack Worth, Juliet Sizmur, Rob Ager & Ben Styles (NFER) – Published June 2015
Questions and discussion