The Northern Ireland Longitudinal Study (NILS) will link the 2021 Census to decades of existing population data, creating a continuous record of people’s lives that stretches back 40 years in Northern Ireland, 30 years in Scotland, and 50 years in England and Wales. This matters because researchers currently lack a single, long-term view of how people’s health, employment, housing, and language use change over time and across the UK. Without this linked data, it is difficult to track the real-world effects of policy decisions—such as how a new housing scheme affects a neighbourhood over decades, or how language policy interacts with economic opportunity. If the project succeeds, it will quietly improve the infrastructure that governments and academics rely on to make evidence-based decisions. The NILS will support research on everything from health outcomes to linguistic studies, and a new Impact Research Assistant will help turn findings into policy-relevant insights. Remote secure access will widen the pool of users, while a cross-UK small-grants scheme will prime new research. The team will also explore using AI to monitor outputs and identify research opportunities, and will help position the NILS within the emerging UK population data system—whether that system relies on a traditional census or linked administrative data.
View original technical description
The Northern Ireland Longitudinal study (NILS) is a member of the family of UK Census Longitudinal Studies (UK CenLSs) and is an important part of the Northern Ireland (NI) and UK data infrastructure for population research. Funded by the ESRC since 2008, this application makes a case for the further funding of user support until 2030. One element – the continued maintenance of the secure environment including help with project applications, output checking, provision of metadata, and the provision of metrics – is necessarily ‘business as usual’. However, the unique environment of the mid 2020s also gives us the opportunity to do new things, responding to fresh opportunities and adapting to a rapidly-changing environment. The linkage of the 2021 Census to the NILS (and other UK CenLSs) is a major opportunity for high-quality impactful research since it is now possible for analysis to span 40 years in NI, 30 years in Scotland and 50 years in England & Wales. This is therefore a central part of our proposal where the NILS takes the lead on a cross-UK small-grants scheme to prime LS research across the UK. Responding to the need for regional/national impact, another innovation is the employment of an Impact Research Assistant (RA) with the remit of supporting research of public and policy interest and building analytical capacity for the future. New technologies shape the context for the bid; for the first time remote secure access to the NILS is a firm prospect and we commit to working with NISRA to ensure that users can access the NILS without using the secure environment; this is key to increasing the geographical reach of the NILS and user numbers. We will also explore the potential use of AI (with colleagues across QUB) to monitor NILS outputs, collect narrative evidence of impacts, and search out research opportunities, sharing lessons and methods where appropriate. Finally, a strand of work – in coordination with colleagues across the UK – will seek to position the NILS in the emerging UK population data where the options for 2031 range from a traditional Census to a linked administrative data system. The unique environment demands agility to take advantage of opportunities, respond to challenges, and to leverage progress to date and we thus place this at the centre of our bid. Continued support for the NILS-RSU will benefit both academic and government researchers requiring access to the NILS databases, as well as users of this research across diverse thematic areas, from health to linguistic studies. Additionally, it will support the development of quantitative capacity for students and early-career researchers.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know