Considerations in the use of local and national data for evaluating innovation in children’s social care

Design/methodology/approach This paper examines the use of data routinely collected by local authorities as part of the evaluation of innovation. Issues entailed are discussed and illustrated through two case studies of evaluations conducted by the research team within the context of children’s social care in England. Purpose We explore the possibilities in using such national, statutory datasets for evaluating change and the challenges of understanding service patterns and outcomes in complex cases when only a limited view can be gained using existing data. Our discussion also explores how methodologies can adapt to evaluation in these circumstances. Findings The quantitative analysis of local authority data can play an important role in evaluating innovation but researchers will need to address challenges related to: selection of a suitable methodology; identifying appropriate comparator data; accessing data and assessing its quality; and sustaining and increasing the value of analytic work beyond the end of the research. Examples are provided of how the two case studies experienced and addressed these challenges. Originality/value The paper discusses some common issues experienced in quasi-experimental approaches to the quantitative evaluation of children’s services which have, until recently, been rarely used in the sector. There are important considerations which are of relevance to researchers, service leads in children’s social care, data and performance leads, and funders of innovation. Implications of the research for policy and practice * Quasi-experimental methods can be beneficial tools for understanding the impact of innovation in children’s services, but researchers should also consider the complexity of children’s social care and the use of mixed and appropriate methods. * Those funding innovative practice should consider the additional burden on those working with data and the related data infrastructure if wishing to document and analyse innovation in a robust way. * Data which may be assumed to be uniform may in fact not be when considered at a multi-area or national level, and further study of the data recording practice of social care professionals is required.