ESRC's activities following the Longitudinal Studies Strategic Review

Following the independent Longitudinal Studies Strategic Review (PDF, 6.7Mb), ESRC is taking forward work to shape our strategy and priorities for data for research, including longitudinal survey data and other types of data.

The Review Panel's major recommendations for future priorities and innovations include:

  • Creation of an administrative data spine from which to sample a new cohort or refresh existing samples
  • A new birth cohort with accelerated longitudinal design
  • Continuation of Understanding Society and an additional ‘transition to adulthood’ sweep of the Millennium Cohort Study
  • Competitive bids to an innovation fund, including an option for the other cohorts to develop innovative bids for this resource
  • Continuation and further development of a longitudinal resource centre (currently CLOSER)

The Review Panel was asked to make recommendations but not to provide fully-fledged plans. Our work will be informed by the Review and will take advantage of developments in technology, data availability, linkage and analytics to provide robust, relevant and representative data on people and patterns of behaviour, to better inform wider policy and practice across sectors.

Professor Anna Vignoles is currently supporting the ESRC's work in this area, including leading a core group of independent academics and other experts to provide advice, for the rest of 2018 initially. We have also started to consult a wider range of experts including academics, policymakers and funders, and will engage with the study scientific lead teams, and with data and research users. We will use opportunities such as relevant conferences and meetings organised by others to seek input.

This work is focused mainly on the areas of the main recommendations above. It will encompass a variety of activities to address needs for information, currently being identified, to inform our thinking.

A key focus of our work is thinking about ways to use administrative data to help with issues of representativeness of the studies' samples both in the sense of knowing who is not in the data and in the sense of actually trying to include hard to reach groups in our studies. Representativeness of the population of interest is important for a wide range of social science research questions, particularly representation of more marginal groups which are often of particular scientific and policy interest. We will scope the feasibility of commissioning the development of an administrative data spine which could be used to sample future cohort studies and from which existing samples might be refreshed and enhanced with additional samples of marginal groups; it could also facilitate data linkage of a wider range of administrative data to new and existing studies.

A second focus is to prioritise innovation in some key areas including data collection methods, expanding the population coverage (for example to include data collection from other family or household members) and representativeness of cohort studies (for example to incorporate supplementary samples of certain populations), and the scope of what data is collected, sample identification, recruitment and retention, and data harmonisation and comparability; this relates both to the existing studies and a potential new cohort.

A high priority is to identify areas where we need to synthesise existing knowledge, such as evidence on the efficacy of different modes of data collection, and areas where more work is needed to develop our understanding through experimentation and testing. Both are needed to inform commissioning of our data infrastructure plans. This development work over the coming months will be crucial to inform assessment of the feasibility of different options and to shape our strategy, funding priorities and decisions.

Other work already underway aligns with this and will also help inform the way forward. One focus is better recording and communicating of the scientific and societal impact of the studies; this is critically important and there is more that needs to be done on this. Work is also progressing on engagement with survey organisations which play a crucial role in longitudinal studies' design and data collection. Meanwhile work is in-hand to secure funding for the next three waves of Understanding Society.

We see collaboration as essential. One area for collaboration is in ensuring that we gain public support for the use of administrative data for research purposes and that we are proactive on the ethics of using such data – we see ESRC having a particular role here – to ensure greater levels of trust. Collaboration is also vital to avoid duplication of effort, for example in the creation of an administrative data spine.