Academy of Social Science response to ESRC Leadership consultation

15 October 2019

Q.1.

  1. The Academy of Social Sciences welcomes the ESRC’s initiative on research leadership. The ‘Evidence Review’ produced by Professor Flinders and Dr Anderson is excellent.  It makes a particular contribution by examining current practice, initiatives, and gaps, and outlining a clear analytic framework for the various dimensions of research leadership.  It includes a sophisticated discussion of the incentives (or lack of them) not only to encourage but proactively to shape a new generation of social science leaders.
  2. The Review is particularly helpful in distinguishing the generic dimensions of leadership (person management, team working and financial management) from those that particularly distinguish research leadership. We believe that this is useful, for it promotes a discussion of the substantive issues in research and working in the larger, interdisciplinary research teams for which social science research leadership is essential.  We believe that this helps underline the extent to which research leadership is not the same thing as generic public sector leadership, but is always underpinned by engagement with research questions and methods and leading teams to address these.
  3. It is perhaps useful to remember that the policy concern about social science research leadership, as compared to academic research leadership more generally, arose in part from concerns about leadership of large social science-led infrastructure projects, such as a new birth cohort. We believe this highlights the special importance of the research leadership in relation to interdisciplinary projects, which work across various social science disciplines, as well as across social science and biological or physical sciences.  We would also, however, note that there are many successful examples of social science research leadership – in the sense of running large projects, consulting widely across academic disciplines, and engaging with data collection, data ethics, data archiving (to make data available to a wide range of researchers), and data analysis – already in existence.
  4. We agree with the Review in that many elements of research leadership, especially generic elements, could, at least in theory, be provided during early career stages, though as the evidence review points out, even this work could benefit from some strategic impetus. At early stages in research careers, we believe researchers who might work on the larger, more interdisciplinary research questions should be given more opportunities for placements and short-courses that will help them gain a deeper understanding of the substantive research challenges they need to address, many of which require interdisciplinary engagement and multidisciplinary skills, as well as project management (financial and people management) skills.  This would undoubtedly be useful even if many of these early career researchers do not in fact go on to become research leaders of large projects.  In short, it involves thinking of research leadership as building on long-term development.  But it may also mean recognising that while PhDs and post-PhD programmes may be the start of this pathway for many, others may be working in the research sector outside academe and then come into universities at later stages of their careers.  While some of the skills needed are undoubtedly generic leadership and management skills (especially people and financial management), others are particular to substantive research questions.
  1. For instance, it would be important for social scientists to engage with substantive issues of how social science, for instance, interacts with biology or genetics (in the case of cohort studies) to build the sound interdisciplinary base on which research leadership will be built. (For instance, for the importance of these sorts of issues on interdisciplinary work, see Callard, F. and Fitzgerald, D., 2015: Rethinking Interdisciplinarity across the Social Sciences, Palgrave Macmillan; Lyall, C., 2019: Being an Interdisciplinary Academic: How Institutions Shape University Careers, Palgrave Macmillan.)  We suspect another element needed to build interdisciplinary research careers requires greater investment in number and data skills, as many of the disciplines with which social sciences interact require continuing attention to issues of quantification, which are in fact becoming more important as data science and inductive techniques become more important.

Q.2 and Q.3

  1. We should start by noting that while we agree further strategic efforts to promote leadership knowledge and skills would be worthwhile, this should not be at the expense of existing research funding. There are already signs that the funding available for social science research is under pressure.  Although social scientists have done well in the Global Challenges fund, the amount of funding for medium sized research projects, often high quality empirical research on a range of social science research linked to the nation’s well-being (such as education outcomes and evaluative research) is, as we have argued (for instance, in our work on Industrial Strategy), not receiving the investment that we believe it deserves.  We hope that any funding for an initiative on social science research leadership would be given new money, and not come out of the ESRC’s existing research budget.
  2. That said, a national hub could have merits, particularly in helping co-ordinate initiatives and incentives to address the gaps identified in the report, and possibly to link with other initiatives within UKRI and elsewhere (like AMS’ FLIER programme). But we think such a hub should have a clearer, more defined and more practical remit than is apparent in the consultation document, where little detail of its possible role is given.  It should not be primarily a research institution, nor should it duplicate work, especially at the early stages, that is already best placed in, for instance, doctoral training partnerships (DTPs).  The need for more investment in order to train and retain a larger research workforce is an argument already made in relation to the government’s target of achieving investment of 2.4% of GDP for research and development.  Investment from the government’s contribution toward this goal will make the remaining government expenditure on R&D far more effective over the medium term.  The Minister for Universities and Skills has made this argument too.  (See https://figshare.com/articles/21st_Century_PhDs_Why_we_need_better_methods_of_tracking_doctoral_access_experiences_and_outcomes/9917813  for thoughts as to how we can better measure the pathways to retention of post-PhD workforce).
  3. A further question would be how to ensure than any ESRC initiative promotes the sort of joint working with UKRI initiatives that is a bedrock of the new system. Planning joint leadership initiatives may be a way forward.  So while we would view any national hub as helping plan (and possibly administering) initiatives for research leadership training for mid-career and beyond, and perhaps auditing or building on existing early stage training, whether at DTPs or in employing institutions, we believe much more detail about what it might do and how it might work would be needed to ensure that it was effective.
  4. At the early stages of research careers (where identifying potential leaders may be difficult), we think research engagement with other disciplines, policy makers and questions requiring interdisciplinary work is a priority. For instance, in our work with the Market Research Society, we held a series of roundtables for existing cohorts and longitudinal studies about the challenges (and finances) of data collection and sampling, to feed into the ESRC’s consultation on the future of longitudinal studies.  Participants included policy makers and researchers from the Office of National Statistics and various government departments.  From those roundtables it was clear that social science researchers who wanted to become research leaders of these studies needed much deeper engagement with data collection issues, and that expertise about this often resided outside academe, in research institutes or commercial firms.  We argued that setting up a programme of funded placements would be a way of addressing the shortage of data collection knowledge and skills among social scientists who could go on to lead new and existing studies.
  5. One way forward might be if existing DTPs and Impact Acceleration Accounts (IAAs) had additional funds to invest in leadership initiatives (with annual reports back to ESRC/UKRI). These initiatives could even be opened to early years researchers working outside academe (since there is increasing evidence of movement between sectors even in social science research), to expand skills and training.  Concrete programmes could be devised by some of the major ESRC supported research centres and infrastructures to help provide short courses, placements and so on with real substantive content (again, this would need funding).  They could include funding to help prepare for ‘impact’, for instance by preparing for parliamentary select committees or other briefings for policy-makers.  Having a fund to develop this kind of training, open to individuals outside the centres and with regard to the equality, diversity and inclusion obligations of these programmes, could give a real kick-start to research leadership in areas such as longitudinal and cohort studies, various data infrastructures, research centres working on obesity or health-related issues, centres working on environmental change (where thinking is required about systemic and social change, individual behaviour change, economics, and understanding various developments in the natural and physical sciences).   In this way, a national centre could help promote development of pathways and programmes for all stages of research pipeline.  For instance, in early years it could encourage project management and team working skills, and audit spending on short courses, substantive summer programmes and other initiatives to promote interdisciplinarity in concrete and relevant ways.  A special focus could be how to improve number and data skills, currently strongest in economics and less widely spread in other social science disciplines.
  6. Indeed, if this approach were taken, we think these ESRC research centres and infrastructures could also work with a national hub to help develop the sorts of research leadership (as opposed to training) that the Evidence Review argues so persuasively are especially needed at mid-career and early senior level. This model would have several advantages.
  7. First, it would involve centres already grappling with substantive interdisciplinary issues and policy challenges in helping set out the kinds of substantive research leadership that are needed. Some of these are clear:  for instance, in the case of the birth and child cohorts, understanding genetics, and current thinking in biology and child development would benefit the building of interdisciplinary research (and it could even include joint work with biologists considering social structure, education research and other elements of social science).  This could aide in the development of substantively sophisticated research, and help build cohorts of the type the evidence synthesis argues is needed.  Centres are also well-placed to help develop leadership programmes, as the Academy of Medical Sciences, FLIER programme does, but to do so with an eye to the concrete methodological, substantive and policy issues that their particular sort of research requires.  Given the breadth of the social sciences and the issues they engage with, we do not believe a single programme or model would be as successful in promoting context-specific research leadership based on centres of excellence.  This would avoid taking a generic approach, or leaving it to individuals to arrange programmes, mentors and so on.

Q.4    

  1. Our response already proposes working with established centres and infrastructure projects to develop approaches for training for early career staff, and a more advanced research leadership programme for mid-level staff, whose potential for senior leadership is more easily judged. This would of course require a period of liaison with such centres, and an assessment of any gaps in substantive research areas that are urgent priorities.
  2. It is important to repeat that, given the pressure on existing research and training funds for the ESRC (for all sorts of research), we think that new research funding would have to be found for any meaningfully sized programme of research leadership. Done well, this kind of initiative requires a thoughtful programme (as with the FLIER programme), and will incur costs for the ‘suppliers’ of training and placements (whether for early years or mid-career researchers) as well as costs for the recipients.  Seeking some sort of contribution from recipients’ home institutions could help not only in selection competitions but in reducing costs somewhat.  The liaison already in place with the Academy of Medical Sciences could provide useful information about scope and scale.
  3. The Academy of Social Sciences could help in convening roundtables between centres (as already did with MRS, SRA and ESRC in learning lessons about data collection for future cohort and longitudinal research), and moderating discussions to help develop various models of research leadership programmes, drawing on our Fellowship for particularly suitable participants.  We would be very glad to do so.