Rebekah Wilson, Tom Benton, David Teeman, Tamsin Chamberlain
01 February 2007
Working with the DfES, NFER has developed a new model for local authorities to compare and assess their performance across children’s services with their statistical neighbours.
Important features of the model are as follows.
- It focuses exclusively on the indicators defined in the Every Child Matters outcomes framework.
- It is based on sophisticated statistical analysis that explored a large number of potential data sources to choose the best variables to use to define statistical neighbours and the amount of weight each chosen variables are given. These choices were made to optimise the accuracy of the benchmarks provided by the model.
- It has been developed by extensive research into the opinions and needs of stakeholders as well as a review of relevant academic literature.
- It is based on the most up to date information available during the period of development (up to September 2006).
Given the new focus on ECM outcomes it is to be expected that the statistical neighbours defined by the new model differ from the previous Ofsted model and the comparator councils used by the Commission for Social Care Inspection (CSCI). Having said this there is a degree of consistency between the new model and previous ones. Of the ten new statistical neighbours defined for each local authority there are on average between four and five that are in common with the previous Ofsted model and between five and six in common with CSCI’s comparator councils.
NFER held three regional training workshops to introduce this new model. Slides from presentations made at the workshop are available below.