• Space and furnishings
• Personal care routines
• Language reasoning
• Activities
• Interaction
• Programme structure
• Parents and staffing
The Caregiver Interaction Scale (CIS) assessed: positive relationships, permissiveness,
detachment and punitiveness of the main pre-school worker. In order that the more educational
aspects of English centres could be assessed, EPPE researchers developed four additional
ECERS sub-scales; ECERS-Extension (Sylva, Siraj-Blatchford and Taggart, 2003) describing
educational provision in terms of: Language, Mathematics, Science and the Environment, and
Diversity.
Setting the centres in context
At interview centre managers were asked about their links to local authority policy and training
initiatives and services. Senior local authority officers from both Education and Social Services
were also interviewed to find out how each local authority implemented Government early years
policy, especially the Early Years Development Plans that were established to promote
education and care partnerships across different kinds of providers (Siraj-Blatchford et al, 1999).
Case Studies
In addition to the range of quantitative data collected about children, their families and their pre-
school centres, detailed qualitative data was collected using case studies of a range of settings
(‘good’ to ‘effective’) chosen retrospectively on the basis of the multilevel analyses of intake and
outcome measures over the pre-school period. This added the fine-grained detail about how
processes within centres articulate, establish and maintain good practice (Siraj-Blatchford et al,
2003).
The methodology of the EPPE project is thus mixed (combining both qualitative and quantitative
data). These detailed case studies used a variety of methods of data gathering, including
documentary analysis, interviews and observations to illuminate the characteristics of more
successful pre-school centres and assist in the generation of guidance on good practice.
Particular attention was paid to parent involvement, teaching and learning processes, child-adult
interaction and social factors in learning. Inevitably there are difficulties associated with the
retrospective study of process characteristics of centres identified as more or less effective after
children in the EPPE sample have transferred to school, field notes and pre-school centre
histories were conducted to establish the extent of change during the study period. Only settings
whose ECERS-E scores had not changed in the last 2 years were selected as case studies.
Analytic Strategy
The EPPE research was designed to enable the linking of three main sets of data: (1)
information about children's attainment and development (at different points in time), (2)
information about children's personal, social and family characteristics (e.g. age, gender, socio-
economic status [SES] etc), and (3) information about pre-school experience (type of centre and
its characteristics).
Identifying individual centre effects and type of provision at entry to school
Longitudinal research is essential to enable the impact of child characteristics (personal, social
and family) to be disentangled from any influence related to the particular pre-school centre
attended. Multilevel models investigate the clustered nature of the child sample, children being
nested within centres and centres within regions. The first phase of the analysis adopted these
levels in models that attempt to identify any pre-school centre effects at entry to reception class.
Given the disparate nature of children's pre-school experience it was vital to ensure that the
influences of age at assessment, amount and length of pre-school experience and pre-school
7