Section Three: Design of Study
There are two major ways to establish the effects of early education and care on children;
randomised controlled trials such as the Perry Pre-school Project and educational effectiveness
designs such as EPPE. Although the former has been admired for decades for its internal
validity, the EPPE team opted for a value added, longitudinal cohort study because of its
generalisability across regions in the country, across social class and home language, and its
capacity to describe the effects of a range of Early Years provision, e.g., ordinary playgroups,
nurseries and rapidly expanding integrated centres. The team based their research design on
the British tradition of ‘school effectiveness’ studies (Sammons, 1996) that took as their central
question -
What is the contribution of Pre-school/School X to the development of children who attend it
(after taking into account familial and other background factors)?
An answer to this question requires assessment of the development of children followed between
the ages of 3 and 7 years and statistical control for background influences. Initially 114 centres
from four types of provision were selected for the study, but in September 1998 an extension to
the main study was implemented to include nursery schools and the UK’s newest and most
innovative forms of provision; ‘integrated centres’ (centres that combined education and care).
Approximately 3,000 children were recruited overall to the study over the period January 1997 to
April 1999 from 141 pre-school centres.
The EPPE project was designed to study three issues that have important implications for policy
and practice:
• the effects of sessional pre-school education and care on children in the age range 3 - 5;
• the ‘structural’ (e.g. staffing profiles) and ‘process’ characteristics (e.g. interaction styles)
of more effective pre-school centres; and
• the contribution of child and family characteristics to children’s development.
The educational effectiveness design enabled the research team to investigate the progress and
development of individual children (including the impact of personal, socio-economic and family
characteristics), and the effect of individual pre-school centres on children's outcomes at both
entry to school (aged 4+) and at the end of Key Stage 1 (age 7+). Such research designs are
well suited to social and educational research with an institutional focus (Paterson & Goldstein,
1991). The growing field of school effectiveness research has developed an appropriate
methodology for the separation of intake and school influences on children's progress using so
called 'value added' multilevel models (Goldstein, 1987 1995). Prior to the EPPE study such
techniques had not been applied to the pre-school sector, although examples of value added
research for younger ages have been provided (Tymms et al., 1997; Sammons & Smees 1998;
Jesson et al., 1997; Strand, 1997; and Yang & Goldstein, 1997) at the time when EPPE was
being designed.
School effectiveness research during the 1970s and 1980s addressed the question "Does the
particular school attended by a child make a difference?" (Mortimore et al., 1988; Tizard et al.,
1988). The question of internal variations in effectiveness, teacher/class level variations and
stability in effects of particular schools over time assumed importance during the 1990s (e.g.
Luyten 1994; 1995; Hill & Rowe 1996; Sammons 1996). This is the first research to examine the
impact of individual pre-school centres using multilevel approaches. The EPPE project is
designed to examine both the impact of type of pre-school provision as well as allow the
identification of particular pre-school characteristics that have longer-term effects. It is also
designed to establish whether there are differences in the effects of individual pre-school centres
on children's progress and development. In addition, the project explores the impact of pre-
school provision for different groups of children and the extent to which pre-schools are effective
in promoting different kinds of outcomes (cognitive and social/behavioural).