Mobility and distance decay at the aggregated and individual level

auteur Stijn Van Daele
tijdschrift GofS (ISSN: )
jaargang 2010
aflevering Safety, Societal Problems and Citizens' Perceptions. New Empirical Data, Theories and Analyses
onderdeel Artikelen
publicatie datum 24 februari 2010
taal English
pagina 41
samenvatting

The primary concern of this paper is, therefore, whether the number of offences reduces over longer distances on both the aggregate and the individual offender level. As such, this paper contributes to a discussion about the application of the distance decay pattern (Rengert, Piquero, & Jones, 1999; Smith,
Bond, & Townsley, 2009; Van Koppen & De Keijser, 1997) and to a better understanding of the distance decay pattern. It does so by splitting distance decay in its two components:
1. near-home offending: do offenders primarily operate within the vicinity of their
homes, as many studies have found? (for a literature review, see Canter & Youngs,
2008a, pp. 4-6)
2. gradual decline: is there an even or an uneven distribution? Does the proportion of
crimes wane with distance? (Canter & Youngs, 2008a, pp. 7-8)

Using mobility features from criminological literature, we compared mobility in terms of average distance travelled, with the mobility shown by specific types of offender at the aggregate level. We expected more emphasis to be put on crimes further from home for these offenders, compared with the general pattern. However, the average distance travelled to commit a crime is only a central measure and does not reveal any information on the actual shape of the distance decay curve. To flesh out the picture, therefore, we first explored whether high mobility actually results in different mobility patterns. In other words, does the distance decay curve of offences/offenders that are related to high mobility also show a deviating distance decay pattern, or is there straightforward decay, with the curve only slightly ‘stretched’ over the longer distance? Second, we bring distance decay down to the level of the individual offender. Distance decay curves are often explored at the aggregated level (which often includes both near-home offending and decay), or central measures are used to calculate offender travelling behaviour, resulting in a loss of some interesting information (Paul Brantingham & Brantingham, 1984, pp. 222, 227). This is often done because listing a number of individual decay curves hampers a clear overall view of the data. In this paper, instead of focusing on central measures, we measure decay itself – i.e. the distribution of distances travelled, independent of the offenders’ mean distances travelled.