Data as Practice: Measuring Outcomes in Australian Poverty Interventions

• In Event: ‘Evidence-making Intervention’: Transforming Implementation Science 1

Abstract

Poor people in rich countries are typically in contact with multiple systems of health and social care. The design and implementation of ‘evidence-based’ interventions for people with multiple needs is a long-standing policy concern that in recent years has been animated by two new fields. On the one hand, the adoption of implementation science in the human services is intended to ensure standardised and measurable practice, and avoid the messiness of local decisions and judgements that are thought to dilute the efficacy of evidence-based interventions. On the other hand, big data and data linkage are described as having the potential to reduce the costs and imperfections associated with evaluation and monitoring data, and provide a comprehensive, even predictive, view of both populations and interventions. Although distinct fields, both implementation science and data linking have been hailed as evidence-led strategies which can improve interventions.
Yet research on the delivery of health and human services shows that assumptions around the status of data and evidence are troubled in practice. Local contexts of implementation are more meaningful than simply risks to implementation fidelity. Data is contested, dynamic, and both evidence and intervention. It is also multiple: the data produced by practitioners for case files of vulnerable families is distinct from that used to report on government expenditure and population outcomes. This paper, which draws on several research projects, adds to the scholarship on stabilised networks and data in practice, and provides new insights on the implementation of health and human services interventions.

Author

• kylie valentine, UNSW