The Geography of Bus Shelter Damage :

This paper offers unique insights into the distribution of damage to bus shelters, in a single case study area, Merseyside (UK). The geography of bus shelter damage is examined in relation to the criminogenic and socio-economic characteristics of its neighbourhood, and the local land use context. The findings suggest that shelter damage is related in a known and predictable way to known characteristics of its neighbourhood, and that shelters in areas with high levels of anti social behaviour and violence against the person are more susceptible to bus shelter damage. Two key factors in the occurrence of bus shelter damage appear to be lack of capable guardianships and the presence of youths. In relation to the influence of land use, the presence of parks, children s play areas and schools (particularly those whose unauthorised truancy levels were above the national average) were positively correlated with shelter damage. By contrast, negative relationships were found between shelter damage and the presence of pubs, clubs, and off-licenses. The implications of these findings for crime prevention are then discussed, alongside some potential avenues for future research. 1 Andrew Newton is a Senior Research Fellow at the Applied Criminology Centre, University of Huddersfield. Email: [email protected] 2 Kate Bowers is a Reader at the Jill Dando Institute for Crime Science, University College London. Email: [email protected] Internet Journal of Criminology © 2007 www.internetjournalofcriminology.com 2 Introduction Previous research has theorised that bus shelters are vulnerable to criminal damage and vandalism (Easteal and Wilson 1990). There have been some studies which examine crime at bus stops in general (Levine et al, 1986, Loukaitou-Sideris, 1999 and Liggett et al., 2003). However, to the authors knowledge, few studies have focused specifically on bus shelter damage. This paper builds on some preliminary analysis of bus shelter damage (Newton, 2007), which demonstrated that damage is concentrated at particular shelters, and hinted that there is a relationship between shelter damage and certain neighbourhood characteristics. However, this earlier analysis used one year of shelter damage and examined older census data (1991), and bi-variate correlations were performed to examine the relationship between shelter damage and neighbourhood characteristics. This study examines three years of data as opposed to one year, and builds in regression analysis and land use data which were not included in the earlier paper. This paper aims to scope the extent of the problem of bus shelter damage in the case study area of Merseyside, UK. Although mainly viewed as low level criminal behaviour it is important to examine this criminal damage for a number of reasons. Firstly, this type of crime is costly to service providers. It is estimated that the cost of this damage for repairs alone in 2003 on Merseyside was approximately £400,000 (Newton, 2004). Secondly it may add to feelings of insecurity amongst public transport users. Waiting at bus stop is an important component of the whole journey approach to public transport (DETR, 1999). The additional parts are walking to or from a stop, and travelling on a bus. However, if any part of the journey is perceived as unsafe then the whole journey may be replaced by a car trip or a decision might be made not to travel. Thus shelter damage may reduce passenger travel and increase feelings of insecurity. Furthermore, research by the then Department of the Environment, Transport and the Regions (DETR, 1998) suggest that reducing fear of crime and disorder on public transport journeys could increase patronage by 3% at peak and 10% at off peak times. Thirdly, shelter damage may add to feelings of insecurity in the area near to the shelter. Wilson and Kelling (1982) discuss the notion of broken windows and how minor signs of decay may suggest a lack of care and enforcement in the community, and Hough (1995) uses multivariate analysis of the British Crime survey to demonstrate that there is a direct link between disorder and levels of concern about crime. Since focusing on bus shelters in particular is a novelty here, this research first aims to establish useful metrics in the definition of the extent of damage. It then examines the relationship between damage and other crime-related incidents, and identifies some factors of the local environment that may correlate with particularly problematic shelters. It then concludes by discussing the implications of the results for the prevention of bus shelter damage. Before embarking on a description of the empirical analysis, it is useful to summarise some of the theoretical explanations concerning the possible cause of shelter damage to ascertain what might be the most likely motivations for this offence, who the likely perpetrators are, and, importantly, where such events may occur. The following sections therefore begin Internet Journal of Criminology © 2007 www.internetjournalofcriminology.com 3 with a definition of different types of criminal damage, and then move on to discuss possible perpetrators and concepts concerning the geographic pattern of the problem. Defining criminal damage If recorded by the police, bus shelter damage is likely to be classified as criminal damage (or vandalism to use an alternative name). The 1971 Criminal Damage Act refers to an act of criminal damage being performed by ...a person who without lawful excuse destroys or damages any property , although in England and Wales vandalism is not a statutory offence. It has a range of classifications in the legal system ranging from criminal damage up to £20 to criminal damage endangering life and arson (Barker and Bridgeman, 1994). It is estimated that 73% of vandalism is never reported to the police and that only 56% of that reported is officially recorded (Mayhew et al, 1993). Moreover, research by Newton (2007) suggested that in 2000 Merseytravel (the Passenger Transport Executive Group (PTEG) responsible for the provision of public transport in Merseyside) recorded 3116 incidents of damage to bus shelters. An analysis of police records found only 8 offences. For this reason it is unwise to rely on police data alone to scope the problem of criminal damage. A further problem with using police data is that it is difficult to define vandalism by type. For example, a useful classification might concentrate on the targets (buildings, bus shelters, cars and so forth) this is hard to discern using police data. It is also difficult to distinguish between accidental and deliberate damage from these records. A particularly useful classification of criminal damage was proposed by Cohen in 1973. This is based on the motivation of offenders to commit acts of vandalism. Wilson and Healey (1987) adapted this classification to distinguish between seven types of vandalism, which are: Acquisitive To acquire money or property, for example, breaking open telephone boxes Tactical The damage is a conscious tactic, a means to achieve some other end, such as breaking a window to be arrested and get a bed in prison Ideological Similar to tactical vandalism, but carried out to further an explicit ideological cause or to deliver a message, for example, chalking slogans on walls Vindictive Damage in order to obtain revenge, for example, breaking school windows to settle a grudge against the head teacher Play Damage in the context of a game; for example who can break the most windows of a house Malicious Internet Journal of Criminology © 2007 www.internetjournalofcriminology.com 4 An expression of rage or frustration which is often directed at symbolic middle class property. It is this type that has the vicious and apparently senseless facade which people find so difficult to understand Innocuous Damage done to property defined by youth as unimportant or of no value Defining perpetrators The classification above is also useful in providing insights concerning the type of offenders who may criminally damage bus shelters. In terms of bus shelter damage, acquisitive damage is unlikely as no financial gain can be met; play (in terms of damage occurring unplanned with a game) is unlikely due to the typical locations of shelters (on the edges of residential estates or on main roads) and vindictive is improbable as the shelter is effectively provided for the public and ownership is therefore difficult to define. Ideological and tactical damage are plausible but also less likely as there are other potential targets that may be more attractive for this (shops, cars, and large advertising boards, for example). Therefore the primary motives are likely to be malicious and innocuous damage. Research by Geason and Wilson (1990) suggests there are two particularly likely candidates for those who perform such of acts of vandalism on public transport and who, it can be inferred, are most likely to also commit criminal damage to bus shelters. These are: (i) Children and Young Persons (aged 5-19) (ii) Persons intoxicated with alcohol leaving pubs and clubs (where damage occurred because alcohol consumption was a contributory factor).