Longitudinal studies need ‘strong use and impact evidence base’

Written by Web News Editor on . Posted in News

An independent review of the UK’s current programme of longitudinal studies has recommended that funding in this area should continue, but that the benefits the research brings should be better captured.

The report, written by an international panel of experts chaired by Professor Pam Davis-Kean, a professor of psychology at the University of Michigan, highlights the value of the UK’s portfolio of longitudinal studies and the importance of the Economic and Social Research Council’s long-term support for them. The panel noted that they are recognised globally and that ‘only longitudinal studies can answer certain questions’.

However, they also acknowledged that the portfolio is expensive - the programme of studies currently costs more than £20 million per year - and the direct benefits of them are sometimes hard to pin down.

The report made several recommendations:

  • Construct an administrative data spine that can be used for new and existing longitudinal studies
  • Commission a new birth cohort with an accelerated longitudinal design and additional funding for transition to adulthood for the Millennium Cohort Study
  • Continued funding for the household panel study
  • Broaden consultations, extend sampling and introduce time-limited funding
  • Invest in data management and sharing
  • Utilise potential of data linkage and harmonisation
  • Expand training and develop a data dashboard for policymakers.

In response to the report, the ESRC has proposed to develop a new ‘UK Population Laboratory’ that will look into data sharing across longitudinal survey data, administrative data, and other sourcesfor research purposes. It also called for better demonstration of the impact of longitudinal studies.‘We know that the studies are used to explore a wide range of interests […] by users across 50 countries,’ it says. ‘We need a strong use and impact evidence base if we are to continue to invest over 10% of our annual budget in longitudinal resources.’

The RSS Social Statistics section hosted a meeting on this topic in October 2017 and contributed to the review (PDF). Its comments were cited twice in the findings, firstly regarding the need for better longitudinal statistical training to analyse the data properly (p43) and also around the need for better data harmonisation (p62) so that the UK’s longitudinal data can be compared with that generated in other countries.

Fiona Steele, a statistics professor at the London School of Economics and RSS fellow, worked on the RSS response. 'The review provides strong support for continuation of the UK’s world-leading cohort studies and household panel study and offers a number of recommendations, including the increased use of data linkage, while improving access to administrative data and providing researchers with information on the matching process,' she says. 'It is also important that we continue to develop UK capacity to analyse longitudinal data so that their full potential can be realised, and the recommendation to review the current state of training provision is an important starting point.'

The RSS Social Statistics section will be holding a session on this topic with the ESRC's Rebecca Fairbairn at our international conference this September in Cardiff. We are also looking at the review of skills which we think needs more investment to make better use of data. Any fellows interested in engaging further with this issue are welcome; please contact our Social Statistics section.

 

Economic and Social Research Council (ESRC)