Evaluating the Linkage of Cafcass and Ministry of Justice Family Court Data within the SAIL Databank: Improving Data Integration for Family Justice Research

Abstract

This study evaluates the quality and usability of linking data from Cafcass England (CAFE) and Cafcass Cymru (CAFW) with the Ministry of Justice (MoJ) Family Court dataset (FACO) within the Secure Anonymised Information Linkage (SAIL) Databank. We assess linkage rates and propose a model to improve data integration by linking cases and multiple case members.

As part of a Nuffield Family Justice Observatory-funded project, we analysed anonymised, population-level administrative data from Cafcass and MoJ’s Data First initiative. We examined linkage rates using Anonymous Linking Fields (ALFs), stratified by jurisdiction, application type, and individual role in proceedings. Binary logistic regression models were used to identify factors affecting linkage success. Additionally, we explored an improved linkage model using court case numbers to connect cases between Cafcass and FACO and linking multiple case members via their ALF/ALF2. All analyses were conducted within the SAIL Databank under approved data governance protocols.

Of individuals in Cafcass datasets, 87% (England) and 84% (Wales) had a valid ALF between 2007 and 2022. Among those, 61% (England) and 78% (Wales) were successfully linked to FACO (FACO records cover the period 2011-2020). Regression models identified factors influencing linkage success, including law type, application year, gender, and role in proceedings. To enhance linkage quality, we suggest an improved model that connects cases between Cafcass and FACO using court case numbers and links multiple individuals within cases via their ALF/ALF2. This approach enhances linkage completeness and accuracy, ensuring a more robust dataset for research and policy analysis.

This study demonstrates the feasibility of linking Cafcass and MoJ Family Court data while highlighting variations in linkage success based on data source and case characteristics. Understanding these patterns informs improvements in administrative data linkage. Our proposed model enhances linkage by connecting cases and multiple case members, strengthening data integration for family justice research and policy development.