Safety in Numbers?

Twenty-first century criminology is increasingly predicated on numbers. Whilst quantitative research is perceived to be “hard data”; the scientific, gold standard, it runs the risk of dehumanising vulnerable people with very little benefit.


Paula Bowles has taught Criminology at the University of Northampton since 2010. Her research interests focus on historical criminology, zemiology, state and institutional violence.


In childhood, I loved numbers, the ability to manipulate, rearrange, reorder, substitute one for another, to create symmetry and yet always end up with an answer. Numbers were as abstract as a jigsaw puzzle, lots of meaningless pieces that, if assembled in the right way, meant that eventually the whole picture would emerge. Along the way the process could go awry, but there was always certainty, always an answer: a solution to the problem. Importantly, that puzzle or equation could be tackled again and again, and provided all the pieces were in order, the solution would be rendered visible once more.

In adult life, my love of numbers has dissipated, primarily because of their application to people. With a global population inexorably heading toward 8 billion, we have to accept that there are an awful lot of us, even so, the relegation of human beings to a mere number is discomfiting. This aversion to numbering people partly stems from academic study of the Nazi Holocaust, which was facilitated by a determination to reduce individual human lives, first to digits, and then to ashes. It also comes from lived experience: in criminology, as in education, there is a plethora of evidence demonstrating that people can, and do change, often in dramatic ways.

Over sixty years ago, Mannheim accused criminologists of creating an ‘almost general impression […] that everything is known in this field,’ suggesting that ‘most of our “knowledge consists of half-baked truths and slogans, of unwarranted generalizations derived from a small body of observations and inadequate samples’ (1955: 133). Furthermore, criminologists such as Stanley Cohen (1988), Nils Christie (1997) and Jock Young (2011) have long recognised the fundamental flaws, inherent in much of what we now consider the bedrock of scientific criminology. They identify how numbers are often used to obfuscate and confuse, pulling readers down a route, whereby those numbers are presented as having their own intrinsic meaning. Such meaning is entirely distinct from the people whose data is being manipulated. Additionally, those numbers are deemed precise scientific and authoritative, having far more sway than any qualitative research predicated on finding meaning in individual lives.

Despite my antipathy to numbers, recently my attention has been drawn to the concept of self-efficacy, in relation to offender desistance, often focused on prisoners. Much of this research appears flawed in the same way as recognised by the eminent criminologists above. Instead of answering what appears be a straightforward question, many of these reports struggle to even define what they are trying to measure, let alone make sense of the measurements. Certainly, Young’s evocative ‘Datasaur,’ with its belly bloated with complex statistical analysis, seems to loom large in such research (2011: 15)

One recent paper which caught my eye, purports to measure self-efficacy within a local prison, HMP Onley, suggesting that a particular programme can improve both mental health and behaviour (Kelley et al., 2017).  This paper, like very many others in Criminology, appears to offer the promise of tackling a deeply engrained historical penal problem.  This article is formatted in the expected manner, contains lots of academic references and appears in a well-respected journal, all of which sounds extremely encouraging. There is no apparent ethical consideration, but the use of academic language, inclusion of 8 hypotheses, as well as the use of a range of different measures (all represented by acronyms) gives a perception of scientific rigour. There are lots of equations, lots of authoritative statements, even some tables.

However, what exactly do we know? I would argue, very little. It is clear that out of 179 prisoners able to take part in the programme, 53 actually completed it, furthermore another 39 made up a control group. From here, the language changes from numbers to percentages to discuss the demographic background of the men involved in the project. In relation to crimes committed and sentences handed down, the paper becomes far vaguer and there is not even the illusion of measurable activity.

Whilst this is but one article, of very many, the repercussions to such research can be profound. The lack of awareness around the pains of incarceration and the reduction of human experiences to quantitative tests as a measure of “self-efficaciousness” is troubling. Furthermore, such a focus implies that individuals have total control over their improvement: if they do not score well on the tests, this can only be due to their inertia, inability or incompetence. By ignoring the carceral experience, any such numbers can only be indicative, as fundamentally, those numbers represent people with their own ideas, fears, worries and behaviours. Discussions around the types of programmes, particularly when based on payment by results, seem to miss this vital point, but of course it means that failure to achieve measurable results can be blamed on individuals and companies. Of course, none of the above denies quantitative data a place within Criminology, but it has to be meaningful, not just a series of bland statements, algebraic equations and tedious charts.

As Christie makes clear, far too many criminologists focus on ‘[l]ong reports of the obvious. Repetitions. Elaborate calculations leading to what we all know’ (Christie, 1997: 13). Instead of rushing to amass quantitative data that looks and is perceived as sophisticated, yet tells us virtually nothing, we need to consider what Criminology is really about. Instead of continuing to churn out criminology that is ‘dull, tedious and intensely empty as to new insights’, Christie insists we look to the roots of our discipline (1997: 13). He insists that criminology is ‘a science based on material from the core areas of drama. Our theories are based on situations of conflict and heroism, danger and catastrophe, abuses and sacrifices’ (Christie, 1997: 13).

As criminologists, we have a duty to be far more critical, taking nothing for granted and avoiding the dissemination of the trivial!


Christie, Nils, (1997), ‘Four Blocks Against Insight: Notes on the Oversocialization of Criminologists,’ Theoretical Criminology, 1, 1: 13-23

Cohen, Stanley, (1988), Against Criminology, (Oxford: Transaction Books)

Kelley, Thomas M., Hollows, Jacqueline, Lambert, Eric G., Savard, Dennis M. and Pransky, Jack, (2017), Teaching Health Versus Treating Illness: The Efficacy of Three Principles Correctional Counseling with People in an English Prison’, International Journal of Offender Therapy and Comparative Criminology, 1-26

Mannheim, Hermann, (1955), Group Problems in Crime and Punishment and Other Studies in Criminology and Criminal Law, (London: Routledge and Kegan Paul Limited)

Young, Jock, (2011), The Criminological Imagination, (London: Polity Press)


Paula Bowles, Senior Lecturer in Criminology,  University of Northampton


Twitter: @paulabowles

Copyright free image: from Pixabay