Case study: Refugee Integration Outcomes (RIO) project

Case study details

Metadata item Details
Owner:Analysis Function Central Team
Who this is for:All government analysts
Contact:ONS.RIO.Cohort.Study@ONS.gov.uk, RefugeeIntegrationOutcomes@HomeOffice.gov.uk

Department

The Office for National Statistics (ONS) and the Home Office (HO)

Situation and action

The Refugee Integration Outcomes (RIO) project is an innovative joint data linkage project between the HO and the ONS aimed at filling the clear need for robust quantitative evidence on long-term integration outcomes of refugees who were granted asylum or resettled in the UK. The project links HO refugee data with other administrative data across government departments to create an anonymised analytical dataset. Insights from this dataset provide unique evidence on integration outcomes for refugees in the UK, providing public value, allowing society and government to better understand refugees and their integration.

The project aims to remove the silos in which data on this topic usually sit. It has effectively brought together the technical expertise of ONS in data linkage with the political and contextual knowledge and insight of the Home Office and other government departments. The RIO pilot study was set up in 2019 and linked HO data to the NHS Personal Demographic Service data.

Findings from the RIO pilot were published in June 2023. You can also read the accompanying RIO methods paper on the ONS website.

Following the success of the pilot, further datasets have been added, and an innovative data linkage process has been put in place. To design the new linkage process, we engaged:

  • HO linguists to better understand Arabic naming conventions
  • ONS linkage experts to pre-process the names and update linkage algorithms built on Western naming conventions

Additionally, the Ministry of Justice’s SPLINK package (an open-source Python package implementing complex probabilistic data linkage methods), embedding of known refugee behaviour into the match keys, and in-depth clerical matching have boosted match rates from around 50% to over 90%.

Outcome, impact, and results

The most recent iteration of the dataset used:

  • HO Borders and Immigration Data
  • ONS Census 2021 data
  • HMRC Pay-As-You-Earn data

The linked dataset is supplied to HO analysts through the ONS’ Secure Research Service to allow for analysis to take place.

Analysis from the linked dataset was published on 26 September, including insights on economic activity, language proficiency, and housing. The results have also been presented at the Census User Conference and Administrative Data Research UK Conference. These findings are filling evidence gaps and helping inform significant decisions at the Home Office in areas including:

  • refugee family reunion policy
  • Local Authority support
  • housing allocation

The findings are also helping the HO to establish the impact of refugees on the UK economy.

Feedback and endorsement

The results have drawn interest from Other Government Departments, and have helped them determine quality issues in their data. They have also received interest from the devolved nations, such as Northern Ireland Statistics and Research Agency (NISRA), who are investigating the feasibility of their own pilot study.

How this work supports the Analysis Function strategy

The RIO project supports the Analysis Function Strategy by:

  • breaking down data silos
  • innovating with advanced linkage techniques
  • building specialist skills
  • providing policy-relevant insights
  • encouraging analysts to work together across government

 

Return to Case Studies