So, were you surprised by the BBC/ONS crime risk calculator?

On 7 September this year, the BBC published a “crime risk calculator”. A product of BBC News and the Office for National Statistics (ONS), users were invited to “Find your personal risk of being a victim”.

The calculator takes a few variables (age, gender, tenure, employment status and postcode) and produces 12-month ‘risks’ of becoming a victim of a few general categories of crime – including violence, robbery, burglary, criminal damage and vehicle-related theft. The risks are presented by area of residence. Local and national figures are compared. BBC journalist Dominic Casciani begins his commentary by asking, “So were you surprised?”

We were. Not by the figures generated, but by the simple and misleading nature of the tool. There is no shortage of analysis which shows that crime risk, as presented, is contingent on many variables not incorporated in the calculations. The variables that are included are also problematic. The focus on aspects of one’s life that are difficult or impossible to change – like age, gender or home ownership – carries the implicit message that there is little a person can do to reduce their risk of crime. But there is ample research that simple security upgrades and lifestyle adaptations can make a difference.

One crucial variable omitted from the risk calculator relates to victimisation within the past one to five years. This is arguably the single most important attribute contributing to risk of crime, and is the attribute which is most useful to know about for the practical enterprise of crime prevention and the allocation of policing resources.^1,2 To give an idea of the consequence of omitting this risk factor, a prior assault on someone in a household consisting of a couple with adult children flags a 44% increase in the risk of subsequent property crime within a one-year recall period. Single young adults have risk raised by some 38% by such a prior assault.³ In short, the effects of prior victimisation are large enough to swamp the risks claimed by the calculator.

The risks of re-victimisation are even greater within shorter periods.^4,5 Once victimised by property crime, a single young adult living in a deprived inner-city area would experience, on average, at least another such crime within the same-year recall period.⁶ The chances of this household being victimised a second time, given the first such experience, are roughly 50/50.⁷

The above examples help to illustrate the importance of accounting for victimisation history when predicting crime risk. Unlike the BBC/ONS tool, these are point estimates obtained from early statistical models of truncated crime victimisation counts that incorporate all available data on individual, household, lifestyle and area of residence characteristics. But even these research-informed risks are not without caveats. To acquire any practical use, they would require replications via hierarchical modelling of the most recent and weighted data, and to be released as lower and upper bounds (i.e., confidence interval estimation) including area of residence variation.

Repeat victimisation is the elephant in the room when considering crime risk (and perceptions of crime). Yet the most recent compilation of crime statistics issued by ONS reflects a lack of attention to the concentration of crime risk: how a disproportionate number of crimes are experienced by the same victims who, moreover, have specific individual, household and/or area of residence characteristics. The concentration of crime risk has remained substantial and even increased despite two decades of falling crime.^8,9 Extensive measurement of this is to be found in the academic literature, but hides in the shadows when it comes to official publications. Its absence from the crime risk calculator is therefore no surprise, though it is disappointing.

About the authors

Andromachi Tseloni is professor of quantitative criminology and leads the Quantitative and Spatial Criminology Group at Nottingham Trent University. An econometrician/social statistician by training, for nearly 30 years she has analysed crime surveys on the extent and patterns of crime victimisation, perceptions, and the crime drop. She is presently Associate Statistical Editor of the Crime Science Journal, and Executive Committee member of the British Society of Criminology and the Nottingham Crime and Drugs Partnership.

Ken Pease is a forensic psychologist by training, and is currently a visiting professor at UCL. He held chairs at the Universities of Manchester and Saskatchewan, acted as head of the Police Research Group at the Home Office, and was a member of the Parole Board for England and Wales. The bulk of his published work over the last 20 years has concerned crime reduction, and he was honoured by a Festschrift entitled Imagination in Crime Prevention. His latest book, with Jason Roach, is Self-Selection Policing.

References and footnotes

1. Pease K. (1998) Repeat Victimisation: Taking Stock. Police Research Series. Paper No.90. London. Home Office. ^
2. Home Office (2010) Crime Reduction Toolkits: Repeat Victimisation. Archived 2010 [available at http://webarchive.nationalarchives.gov.uk/20100408171529/http://www.crimereduction.homeoffice.gov.uk//toolkits/rv0101.htm], accessed 26 September 2017. ^
3. Tseloni, A., Osborn, D.R., Trickett, A. and Pease, K. (2002) Modelling property crime using the British Crime Survey: What have we learned? British Journal of Criminology, 42, 89-108. ^
4. Bowers, K., Johnson, S.D. and Pease, K. (2005) Victimisation and re-victimisation risk, housing type and area: A study of interactions. Crime Prevention and Community Safety: An International Journal, 7(1), 7-17. ^
5. Farrell, G. and Pease, K. (2014) Repeat victimisation. Encyclopedia of Criminology and Criminal Justice (ECCJ), Gerben Bruinsma and David Weisburd (Editors in Chief). New York: Springer-Verlag, pp. 4371-4381. ^
6. This is given by crime concentration, which for this particular household type has been estimated at between 2.16 and 2.90 (see Pease and Tseloni 2014, Tables 4.2 and A.2.5). ^
7. Pease, K. and Tseloni, A. (2014) Using modelling to predict and prevent victimisation. Springer-Brief Criminology Series, New York: Springer. ISBN 9783319031842 (Print) 9783319031859 (Online). ^
8. Ignatans, D. and Pease, K. (2016). On whom does the burden of crime fall now? Changes over time in counts and concentration. International Review of Victimology, 22(1), 55–63. ^
9. Hunter, J. and Tseloni, A. (2016) Equity, justice and the crime drop: The case of burglary in England and Wales. Crime Science. 5(3). DOI: 10.1186/s40163-016-0051-z. Open Access. ^