SOUTHAMPTON UK POPULATION LAB SCOPING PROJECTS

Lead Research Organisation: University of Southampton
Department Name: Sch of Economic, Social & Political Sci

Abstract

The proposed research takes the evidence from published and unpublished work on the design, implementation and analysis of longitudinal studies, and uses it to develop strategies which can be considered for longitudinal data investments in the UK. This includes the maintenance and development of existing sources as well as the potential for development of new ones. It covers modern approaches to supplementing survey data collection with administrative data, and a variety of different methods which provide information for the general population and for specific groups of interest. It balances the risk and complexity of advanced approaches against the need for simplicity to make these data sources available to researchers, and against efficiency. Efficiency is important to make the best use of scarce resources to manage and develop the UK's longitudinal data infrastructure.

Planned Impact

The outputs from these investigations of six areas of survey design will provide recommendations for the design of future longitudinal studies and the analysis of all the ESRC's longitudinal data investments. The direct impact is on the way in which these studies are implemented and used, which is expected to affect the ESRC and the bodies which undertake these surveys. The information should provide information to use funding efficiently, and therefore maximise the effectiveness of investments. Ultimately the research will benefit the users of longitudinal studies - academics and policymakers who use the information to make public services as efficient and effective as possible.
 
Description The series of reports examined different approaches for gathering and analysing longitudinal data using major infrastructure surveys. It highlighted areas for further development/research, including:
• a need to work on the design of a whole system of longitudinal surveys, and how this could be supported by a modular approach.
• a need to understand administrative data that can be used to support longitudinal analysis, and how timing differences in such records affect longitudinal analyses
• We concluded that non-probability sampling methods are not currently suitable for a national longitudinal infrastructure resource. But they may be useful for particular parts of the population. Therefore more research is needed on how to combine probability samples and non-probability data sources (including focused studies using sampling techniques designed for hard to reach populations), and how such combined data resources can be analysed in a principled way accounting for their different types of errors.
Exploitation Route The main areas for further development need to be taken forward by researchers in statistics, social science and government. Some of the review material and guidelines are to be put to use in maintaining existing, and implementing new, longitudinal studies.
Sectors Communities and Social Services/Policy,Government, Democracy and Justice,Other

 
Description The findings have contributed to the ESRC's thinking on the future ways to fund and operate major longitudinal infrastructure projects, particularly through the Longitudinal Studies Workshop (see "engagement activities"). This influence is seen in the Early Life Cohort plans where there is no expectation of the need to use an accelerated design. The findings have also contributed to the knowledge available to the Early Life Cohort Advisory Group (ELC-AG) of which the PI for this award is a member.
First Year Of Impact 2019
Sector Other
Impact Types Policy & public services

 
Description Membership of Early Life Cohort Advisory Group
Geographic Reach National 
Policy Influence Type Membership of a guideline committee
 
Description INGRID SIG workshop on non-probability samples 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Around 30 statistical researchers and practitioners met to exchange ideas on how to work with non-probability data, including some debates about the potential for non-probability sources in social science, official statistics and academic research
Year(s) Of Engagement Activity 2020
URL http://www.inclusivegrowth.eu/special-interest-groups/sig-unitrier-27-28-february-2020
 
Description Longitudinal Studies Innovation Workshop 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Other audiences
Results and Impact Around 40 academics, researchers and ESRC staff attended a workshop to help define areas of innovation and directions of travel for the funding of longitudinal surveys in the UK in the future. There was considerable discussion, which helped to define areas for future research calls.
Year(s) Of Engagement Activity 2019