Some Thoughts on Richard Sparks’ report to the Economic and Social Research Council

Using existing surveys

Richard argues that, as a community, we ought to make more use of existing surveys. As a keen secondary data analyst, I wholeheartedly support this suggestion. However, my call would be that we ought not just to re/use the ‘usual suspects’ (such as the Crime Survey for England & Wales, the European Social Survey, or the various bespoke surveys which have been generated). As part of a small ESRC-funded ‘scoping’ project, I undertook a review of the existing datasets held by bodies such as the UK Data Archive and which could be useful for assessing the impact of Thatcherism on crime, the criminal justice system, popular opinions on crime and so on. I uncovered a whole host of studies which in one way or another touched on matters of interest to criminologists (such as the British Election Study series, the General Household Surveys, The British Social Attitudes Surveys, the World Values series and the European Social Survey). There are of course also the Youth Cohort Study series and the Young People and Crime surveys. So, there are lots of secondary data sets ‘out there’ beyond the confines of the ones which we normally turn to. Not to mention surveys from other countries too.

Furthermore, let us not limit our thinking about using existing surveys to solely the data which have already been produced and lodged for analysis. With an earlier tranche of ESRC funding, Jon Jackson, Emily Gray and I analysed data based on survey questions which David Gadd and I had designed as part of an earlier ESRC award and which colleagues at the Home Office (who then administered the BCS) had agreed to put in the British Crime Survey (as the Crime Survey for England & Wales was then known). The British Crime Survey (BCS) team fielded three questions (each of which had two follow-up questions depending on respondents’ answers). These data were the bedrock of a book which the three of us wrote and which was published by the OUP to much acclaim. One possible future funding stream, building on this model, could include the chance to design (or in some cases, select from other surveys) questions for use in an existing survey. This would allow criminological matters to be addressed more fully than is currently the case via the enabling of topics in, for example, political science to be analysed alongside criminological topics. Why not, for example, ask questions based on time–space budgets from time use surveys in the Crime Survey for England & Wales? Does the nature of the use of time and space shed any light on the reasons for victimisation? Do victims change their daily routines? If so, in what way(s)?

Taking change over time serious via longitudinal studies

Richard also referred to the United Kingdom’s heritage of longitudinal studies. The United Kingdom has a wonderful history of investment in such studies. The Cambridge Study of Delinquent Development, the Peterborough Study and the Edinburgh Study have all (a) been funded by the ESRC (and others in some instances) and (b) made lasting contributions to criminological knowledge and policy development. However, each shares one weakness. All are studies of one locale. In a series of recent papers in the British Journal of Criminology, Geoforum and Politics & Society, Emily Gray, Phil Jones and I explore longitudinal data collected in England, Scotland and Wales (Farrall et al., 2019, 2020a, 2020b). These data sets are studies of samples of entire birth cohorts, meaning that they are not limited to just those attending a certain school or living in a particular area. This may sound like a small difference, but these national studies allowed us to explore variations in economic trajectories. Some parts of the United Kingdom experienced economic and social changes which had great impacts on the lives and life courses of the children who were growing up in them. We could only chart, for example, the part played by economic restructuring across Britain because the data included a wide range of experiences which played out along county and regional lines. By linking to census data from the 1961, 1971 and 1981 censuses, we were able to analyse two such cohorts (one born in 1958 and the other in 1970). By strategically pairing these data in order to illuminate how structural influences themselves are not invariant we were able to show how social and economic change – sometimes the result of government policies, sometimes the drivers of such policies – influenced their life-courses and experiences of crime (see Farrall et al., 2019, 2020a, 2020b). But let us not fall into the trap of thinking that one needs large numbers to explore economic change; qualitative longitudinal research (such as that undertaken by Rod MacDonald, Colin Webster and Tracy Shildrick) can also illuminate such experiences and ought to be part of any future research strategy.

Thinking about funding instruments

Let me turn now to one of the things which I spoke about during the session in Edinburgh when Richard unveiled his report. Reflecting on my own history of research funding from the ESRC, I pointed out that on two occasions I had received ESRC support which today may be referred to as ‘pump-priming’. The first of these (for around £11,000 if I recall correctly) was to design the survey questions which were used in the British Crime Survey and which Jon Jackson, Emily Gray and I then analysed using ESRC support. Prior to this point, I had started to wonder (having failed with all of the previous applications for funding I had made) whether I ought to continue to pursue a career in academia. The questions which I (with input from David Gadd) designed with this ESRC grant were then used (for a few years) in the European Social Survey and various other bespoke surveys both in the United Kingdom and overseas. A couple of years later, my initial interest in Thatcherism also led to, and was bolstered by, a further small grant from the ESRC (this time for around £30,000). This enabled me to review the existing data sets (as I referred to above) and led to two further ESRC awards (one which explored macro-level processes with Colin Hay, Will Jennings and Emily Gray), and one with Emily Gray and Phil Jones to augment these analyses with individual-level longitudinal data (using the 1958 and 1970 birth cohorts, as I mention above). As such, small grants allowed someone (me) with no track record of holding grants a chance to take a step into doing so and then allowed me to pump-prime the exploration of a topic (Thatcherite policy impacts and their empirical assessment) on which I had no prior engagement, but which have led on to around £1 million worth of funding.

Small grants, much to my regret, were culled by the ESRC about 10 years ago. The beauty of them was that they were assessed by the Grant Assessment Panel (and so did not go out for peer review). This made the turn-around of them faster than is the case for grants now, and meant that the ‘wise ol’ heads’ on Grant Assessment Panels could risk a relatively small amount of money on what might turn out to be a novel approach to an old topic, or a new area of research. The move to larger grants, I suspect, has (a) slowed things down generally (since peer review takes time to organise and complete) and (b) (more importantly) led to an understandable conservativism in the awards made. After all, if a scholar’s suggestions take a bit of a gamble on a research topic costing £30,000, this is less risky than if they had asked for £300,000. No funder wants to ‘waste’ money or to back something which puts at risk their legitimacy by backing what proves to be a huge flop. Small grants, along with the seminar series, allow budding scholars (and indeed ‘ol’ heads’ with some great new ideas) to start to explore these at minimum risk to the ESRC and its reputation in the eyes of the UK tax payers who so generously support UKRI. Our ‘crazy ideas’ are tomorrow’s research programmes and while we are not short of ‘crazy ideas’, we need to think carefully about the best ways of funding them so that ECRs (amongst others) are able to take their first steps towards grant success and so that risks are taken with an appropriate degree of acceptable outcomes. This way we can harness all those wonderful ideas which (at present) are not being funded simply because there are few ways of securing pump-priming. Accordingly, I call on the ESRC to find exciting new ways of supporting colleagues who need smaller amounts of resourcing in order to push forward their visions and agendas. After all, little bit of money can go a long, long way. Then can we be confident of a thriving research community in the years to come.