Designing the Swedish Crime Harm Index: an Evidence-Based Strategy

Fredrik Kärrholm,John Smaaland,Peter Neyroud

doi:10.1007/s41887-020-00041-4

Abstract

Research QuestionWhich method for deriving a Crime Harm Index (Policing 10:171–183, 2016) for Sweden from criminal justice sources offers the best evidence for providing a sensitive indicator of differences in harm levels across offence categories?DataThe number of days of imprisonment for each offence category associated with five different kinds of scales were extracted and compared: consensus by an expert panel of judges, the statutory maximum penalties, statutory minimum penalties, the average of maximum and minimum penalties and the average of actual sentences imposed in a recent time period for each crime type. Unlike the UK, for which the Cambridge Crime Harm Index draws on sentencing guidelines, Sweden has no such guidelines to offer.MethodsThe data were compared for sensitivity defined as the difference in length of imprisonment days between high and low severity crimes, as well as other characteristics of the data sources.FindingsGiven the available alternatives, the sentencing data average of actual sentences handed down by crime type provided the greatest reliability and sensitivity across the penalties for offences of high and low severity. Applying that method to both crime trends and crime mapping produces substantially different results from counting all crimes with equal weight and can be used by police and others to allocate resources with greater precision in relation to harm prevention.ConclusionsOn both empirical and normative grounds, the average sentences in a recent time period for each crime category provides the most sensitive and democratic method for establishing an officially recognized Swedish Crime Harm Index.

Highlights

The only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harms to others. (John Stuart Mill ([1859] 1978: p. 9).Different crime types cannot be valued
Statistical tools have been developed in several countries for weighting crime counts according to the severity of different crime types
One important difference is that more than half of the crimes have the same value regarding minimum sentences. This small variability makes a Crime Harm Indices (CHI) with such a basis less useful, as its purpose is to function as quantitative method to recognize differences

Summary

Introduction

Different crime types cannot be valued . In the process of identifying geographical areas that are to be prioritized for patrolling, it is irrational to value serious assault and verbal harassment . We argue that precise and correct decision-making is dependent on systematic statistical methods. For this purpose, statistical tools have been developed in several countries for weighting crime counts according to the severity of different crime types. Statistical tools have been developed in several countries for weighting crime counts according to the severity of different crime types These tools are often referred to as Crime Harm Indices (CHI) (Sherman et al 2016)

Methods

Results

Conclusion