Professor Kathy Sylva University of Oxford

EECERA C

onfe

rence

Tal

linn E

stonia

2013

Quality in Early Childhood Education – Can it be International?

The EPPE/EPPSE Team

Principal Investigators:

Kathy Sylva Department of Education, University of Oxford

Edward Melhuish Birkbeck, University of London

Pam Sammons Department of Education, University of Oxford

Iram Siraj-Blatchford Institute of Education, University of London

Brenda Taggart Institute of Education, University of London

Themes in today’s presentation

The effects of pre-school education on children’s intellectual and social/behavioural development at age 11, 14 years

(Controlling for for) effects of early learning at home

Outcomes – attainment or progress?

Independent Variable: Early Childhood Education

e.g. Perry Pre-school Programme

Dependent Variable: Children’s developmental outcomes

Establishing cause in educational research:Experimental design

An experimental approach:The High/Scope Perry Study in US

School

Achievement Tests,plus

Interviews with children

Social Profiles:questionnaires

Interviews with parents

Behaviour Ratings:teachers

Assignment to Special Education

I.Q.

Jobs

Training

Parenthood

Welfare

Crime

Pre

-sch

ool g

roup

Con

trol

(no

sch

ool)

grou

p

3 5 AGE 18 27

From Schweinhart & Weikart (1993)

An educational effectiveness approach: fixed effects in EPPE

The EPPE Project uses an educational /school effectiveness design, which investigates ‘natural’ variation using multi-level modelling (pupil and school level) in order to establish effects of pre-school attendance or quality on children’s development over time.

Five regions strategically selected in England (

141 pre-school centres randomly selected within regions to include: playgroups, nursery classes, private day nurseries, day care centres run by local authorities, nursery schools and integrated centres

2800 randomly selected children from 141 centres; 305 children from home

Linked study of 800 children in Northern Ireland; ‘strength of replication’

The EPPE Sample

EPPE has an ‘mixed methods’ design which includes:

Quantitative methods (led by Sammons)

Qualitative methods (led by Siraj-Blatchford)

50 cast studies of resilient chidren and families

The EPPE Mixed Method Design

25 nursery classes

590 children

34 playgroups

610 children

31 private day nurseries

520 children

20 nursery schools

520 children

7 integrated centres

190 children

24 local authority day care nurseries

430 children

home

310 children

Design of EPPE 3-11: 6 LA, 141 pre-schools, 3,000 children

Reception Yr 2Pre-school Provision (3+ yrs)

Key Stage 1

600 Schools

Yr 6

Key Stage 2

1000 Schools

Neighbourhood

The Developing

Child

Family

Settings

Cultural context

Immediate environment

Social and economic context

Bronfenbrenner’s model of human development

Influences on child outcomes at ages 5, 7, 11, 14

Child Factors

Family Factors

Home-Learning- Environment

Cognitive outcomes:English & maths

Social/Behavioural:Self RegulationLikes to work things out for selfPro-socialConsiderate of others feelingsHyperactivityRestless, cannot stay still for longAnti-social Bullies other children

Primary School

Pre-School

Secondary School

Sources of data

Child cog/ social/dispositional assessments

Child/Family background information, e.g. SES, birthweight

Interviews with all parents

Case studies of settings and of children who were ‘resilient’

Observation rating scales for quality (ECERS-R and E)

Early Years Home Learning Environment (HLE at ages 3-4 years)

HLE index (Melhuish, 2001), measuring frequency of:

1. Reading with child2. Painting and drawing3. Library visits4. Playing with letters/numbers5. Teaching alphabet6. Playing or teaching numbers/shapes7. Playing with songs/nursery

rhymes

HLE to children’s British Ability Scales scores as well as measures of socio-economic status and educational qualifications of parents.

0.21 0.22

0.40

0.57

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

14-19 20-24 25-32 33-45

Net Effect of Early Home-Learning Environment on Maths at Age 10Effect Size

HLE Index (Pre-School)

R

Multilevel analysis predicting oral language skills at age 5 (primary school entry)*

Factors Effect

SizeNo. of Siblings (3+ vs. none)

-.17

English as an Additional Language (EAL vs. English as mother tongue)

-.24

Mother’s Qualification Level (degree vs. no degree)

.24

Family SES (semi-skilled manual vs. professional non-manual)

-.23

HLE: Being read to (daily vs. rarely)

.27

Effects of Early Years HLE on vocabulary at 5

Does pre-school attendance matter at 7 years?

READING at key stage 1, social class and pre-school experience

WRITING at key stage 1, social class and pre-school experience

The effect of social class and pre-school attendance on literacy at age 7

1.8

2.0

2.2

2.4

2.6

2.8

Professional Skilled Un/semi skilled

Social class by occupation

Me

an

ye

ar

2 r

ea

din

g le

vel

Pre-school

Expected minimum

No pre-school

1.6

1.8

2.0

2.2

2.4

2.6

Professional Skilled Un/semi skilled

Social class by occupation

Me

an

ye

ar

2 w

ritin

g le

vel

Pre-school

Expected minimum

No pre-school

From Sylva et al. (2004)

Does pre-school quality matter at age 11?

ECERS-R

Based on observation 7 sub-scales:

Space and furnishings Personal care routines Language reasoning Activities Interaction Programme structure Parents and staff

Harms, Clifford & Cryer (1998)

ECERS-E

Based on observation 4 sub-scales:

Literacy

Mathematics

Science and environment

Diversity

Sylva, Siraj-Blatchford & Taggart (2010)

Two Early Childhood Environment Quality Rating Scales

Example ECERS-E item: Food preparation (Science)

Inadequate Minimal Good Excellent

1 2 3 4 5 6 7

1.1 No preparation of food/drink is undertaken in front of children

3.1 Food preparation is undertaken by staff in front of the children

5.1 Food preparation / cooking activities are provided regularly

7.1 A variety of cooking activities in which all children may take part are provided regularly

3.2 Some children can choose to participate in food preparation

5.2 Most of the children have the opportunity to participate in food preparation

7.2 The ingredients are attractive and the end result is edible and appreciated

3.3 Some food-related discussion takes place where appropriate

5.3 The staff lead discussion about the food involved and use appropriate language

7.3 The staff lead and encourage discussion on the process of food preparation and/or question children about it

5.4 Children are encouraged to use more than one sense (feel, smell, taste) to explore raw ingredients

0

1

2

3

4

5

Literacy Mathematics Science and environment Diversity

EC

ER

S-E

sco

re

Level 2 Level 3 / 4 Level 5

ECERS-E subscales by manager qualifications

The effect of pre-school pedagogical quality (ECERS-E) on English and Mathematics in Year 6

There is no effect of the ECERS-R on English or Mathematics at age 11

The effect of global quality (ECERS-R) on academic attainment

The social behavioural outcomes(Goodman scale extended)

Hyperactivitye.g. Restless, overactive, cannot stay still for long’

Self regulatione.g. ‘Likes to work out things for self’

Pro-social behavioure.g. ‘Considerate of other people’s feelings’

Antisocial behavioure.g. ‘Bullies other children`

The effect of pre-school quality (ECERS-R and ECERS-E) on self-regulation at age 11

The effect of pre-school quality (ECERS-R and ECERS-E) on anti-social behaviour at age 11

The net effect of pre-school quality on academic progress in English between 7 and 11 (accelerated learning

compared to peers)

Representation of mediated effects upon literacy and numeracy

Early Years HLE3+

Pre-school Effectiveness3-5 years

Literacyages 7, 11, 14

Numeracyages 7, 11, 14

Self-regulationat age 5

Quality fosters capacity for learning how to learn

Pre-school attendance alone led to better attainment in English and Mathematics but this did not translate into better progress.

Pre-school quality (medium, high) led to greater progress between Years 2 and 6 for English and Maths, children who attended high quality had accelerated learning compared to ‘home’ children.

High quality pre-school provides children with an initial boost to at school entry but also promotes progress (by fostering children’s capacity to learn?)

Play scenario: KatzFour children were playing together. Three were wearing trainers but shoes of one child lit up occasionally.

Teacher: Wow! Look at your shoes! That is so cool. They light up when you step down. Child 1: Yes, they do this. [Jumps up and down several times]Teacher: How does that happen? How does it light up?Child 1: Because they are new.Teacher: Um. Mine are new too but they don’t light up. Child 2: No, because they light up when you step down on them. [Steps down hard several times]

Quality learning: The Light-Up Shoes

Teacher: [Steps down hard several times] That’s funny. Mine don’t light up when I step down.Child 3: No, no, no, you have to have these holes [points to the holes]Teacher: [Pointing to the holes in her own shoe] But I have holes and mine still don’t light up, and Josh has holes in his trainers too and his do not light up either. I wonder why?Child 4: I think you need batteries. Kids, you need batteries. Child 1: Yeah, you need batteries to make them work. [Thinks for a while]. But I did not see batteries when I put my toes in. Child 4: I think they are under the toes. Child 2: I can’t feel the batteries under my toes. Teacher: I wonder how we can find out about this?

The Light-Up Shoes (cont.)

Harms. T., Clifford, R. M., & Cryer, D. (1998). Early Childhood Environment Rating Scale Revised Edition (ECERS-R). New York: Teachers’ College Press.

Sylva, K., Melhuish, E., Sammons, P., Siraj-Blatchford, I., & Taggart, B. (2010). Early Childhood Matters: Evidence from the Effective Pre-school and Primary Education Project. Oxford: Routledge.

Sylva, K., Siraj-Blatchford, I., & Taggart, B. (2003, Second Edition 2010). Assessing quality in the early years. Trentham Books.

EPPSE Projecthttp://eppe.ioe.ac.uk/

Families, Early Learning and Literacy (FELL) research grouphttp://www.education.ox.ac.uk/research/fell/