Towards a standardised approach in cross-national spatial crime research : a case study analysis

Abstract

The lack of a standardised approach in cross-national spatial crime research limits the creation of consistent, reliable and valid knowledge regarding the spatial dimension of crime and crime causation across contexts. This begs the question: what is the most appropriate or ‘correct’ methodological approach to follow in order to measure crime and its causation in any context? This question is initially answered by undertaking a spatial analysis of crime in Khayelitsha, South Africa. Khayeltisha (the primary study site) is home to one of the most deprived, marginalised, crime-ridden and socially disorganised communities, not only in South Africa but in the world. A largely ineffective police force together with unchecked and informal policing structures, unreliable and uncertain political support have contributed to unmanageable crime. Subsequently, there is a desperate need to generate valid empirical knowledge and an understanding of crime patterns as they relate to the social and economic environment. This study uses social disorganisation as a guiding framework to gain a more thorough theoretical understanding of crime and the ecological characteristics that can affect it within the context of Khayelitsha. The central tenets of the theory, that describe the level of social disorganisation of a place, include socio-economic deprivation, family disruption, residential mobility, racial heterogeneity and urbanisation. These characteristics are however notoriously difficult to quantify and the choice of variables used to represent the central tenets of the theory are often inconsistently interpreted. In the first part of the study, I demonstrate how the numerous ways in which these characteristics can be operationalised continuously yield conflicting results, which raises questions regarding its applicability in a developing context. In the analysis, three separate sets of variables, consistent with social disorganisation theory, are tested using regression modelling, to illustrate these inconsistencies. Results are shown to vary in significance and predicted associations across almost all the individual variables representing the same structural neighbourhood characteristics. This raises the question of appropriate definitions and descriptions of variables together with the suitability of various combinations of variables used as proxies for concepts synonymous with social disorganisation theory. The findings also refute any assumption that different sets of variables applied as proxies for the same tenets of social disorganisation would perform similarly and bring into question results from the plethora of prior local and international research that have used the social disorganisation theory as a framework to guide the selection of variables in spatial crime analysis. As a solution, in the second part of the study, I propose the use of standardised variables in spatial crime research using variables informed by the International Organization for Standardization (ISO) indicators. These variables are used to measure social disorganisation in Khayelitsha (in South Africa) and Fort Lauderdale (the secondary and comparative study site in the United States), allowing for a cross-national comparison of the causes of crime in these two contexts. The ISO 37120 indicators, in particular, provide guidelines to create clearly defined, standardised measurements, and consequently consistent indicators comparable over time or space. This is currently lacking in cross-national comparative crime studies applying social disorganisation theory. The indicators promote cross-national information sharing, transparency and open data. Furthermore, this approach offers the benefit of contributing to the creation of a reliable foundation of globally standardised data, which can assist in building core knowledge and can be used for comparative knowledge sharing in spatial research of crime.