Is there a link between design and crime.

Designing crime out of our public realm?

Crime plays a key role in influencing the experiences and positive sustainability of the built environment – where crime is high, this can create community concern and misery; where it is low, the neighbourhood is typically viewed in a positive light. Related to these crime levels is the design of the neighbourhood, with some types of housing and street network design leading to different levels of opportunity for crime to occur.

At present, there is no clear and consistent official guidance on designing out crime for urban environments that can be used to encourage urban designers to consider crime prevention issues in the formative stages of their work. In addition, there is a lack of evidence on the impact and influence that contemporary higher density housing developments and their design attributes have on crime, making it difficult to relate existing crime prevention guidance to the design of these types of developments. There is a need to produce this evidence base in order to inform new guidance, illustrated with examples of good practice, in order to provide designers with practical advice that can help them to minimise the crime experience of new high density housing developments.

Crime research consistently demonstrates that crime is spatially concentrated at a variety of different scales. Furthermore, while crime can be socio-economical sensitive, trying to draw direct correlations between community indices and place is at the onset fraught with difficulty. Drawing on a wealth of crime data, and accepting that crime may be driven by opportunity rather than absolute intent, research undertaken at a micro-level has been able to identify patterns of repeat burglary; repeat victimisation consistently shows that victimisation is a good predictor of future risk; that the timing of re-victimisation has a distinct periodicity and a short half-life, and that repeat victimisation is most prevalent in high-crime areas.

Two theoretical processes have been proposed to explain patterns of repeat victimisation. The first suggests that repeat victimisation is the consequence of a contagion-like process. If a home has been burgled on one occasion, the risk to the home is boosted, most likely because offenders will return to exploit good opportunities further (e.g. to steal replaced items or those left behind). In contrast, the second suggests that repeat victimisation may be explained by time-stable variation in risk across homes and a chance process. Different offenders independently target attractive locations for which risk is flagged. Understanding the contribution of the two explanations is important for both criminological understanding and crime reduction. Hitherto, research concerned with repeat victimisation has adopted a top-down methodology, analysing either victimisation or offender data. Research undertaken by UCL suggest that increasing the heterogeneity of target attractiveness can generate spatial concentrations of crime not dissimilar to those discussed above, but that a contagion-like process is (also) required to generate the time course of repeat victimisation.

The issue most germane for recorded crime data is that of under-reporting. Many crimes are simply not reported to the police. For some types of crime, this has implications that require no articulation. However, in the UK (for example), burglary is one of the most frequently reported crimes, with around 85% of burglaries with loss being reported. Thus, recorded crime data may (arguably) be considered as a reasonable proxy for patterns of victimisation, and, where particular signatures are considered (rather than general trends in volume for example), problems associated with under-reporting are substantially reduced. A more specific limitation with these types of data is that the accurate quantification of how risks vary within different areas is difficult. The extent to which this is a problem depends on the size of the area considered. For large areas, risk is likely to vary considerably across homes, meaning that analyses conducted using data aggregated at the macro-level will underestimate the variation in risk across units. With increasingly smaller units of analysis, however, characteristics such as architectural design, levels of natural surveillance, access and escape routes, and, consequently, crime risks, will become more homogeneous across homes. Thus, analyses conducted at the micro-level of analysis may minimise the above concern and provide a useful estimate of risk heterogeneity across homes. This is particularly likely to be the case where, for the unit of analysis selected, areas have been defined so as to maximise homogeneity within areas, whilst maximising observed differences between them.

The on-going debate between “open” environments and “closed” environments are driven predominantly by the design debate; while numerous studies have been undertaken to prove the validity or not of these positions, because collected crime data is so very varied and at times ambiguous, subjective interpretation of data has been able to motivate for both positions.

Furthermore, the nature of crime within newer settlements (say constructed over the last ten years) will be atypical and may change over time to reflect the crime averages of the surrounding, more established areas. While the socio-economic profile of these newer communities may statistically be on parity with the surrounding areas, those essential community linkages and human interactions may not yet have formed (i.e. the idea of neighbourliness does not yet exist), hence either creating greater opportunity for crime or, due to better design and installing crime prevention measure to the new units, may reduce crime. The exact causation therefore between crime and design is therefore unresolved.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s