The current context on big data in criminology and criminal justice

If organisational culture and the principles and processes underlying big data and data analytics are some of the key challenges for organisations in general, they are especially relevant for criminology and criminal justice. Criminology and criminal justice are largely considered applied disciplines where data collected through research are used, at least in theory, to inform policy and practical decisions (Lynch, 2018). The applied dimension of criminology typically involves the analysis of data in the form of case studies, prevalence rates, trends, and statistical associations for the purposes of informing law enforcement, the courts, and corrections as to evidence-based best practices. To these ends, both the discipline of criminology and the criminal justice systems that it seeks to inform have begun to integrate, to varying degrees, big data and data analytics. These are new developments in criminology, and undoubtedly are not being utilised to the same degree as in commercial contexts, such as longstanding practices of tailored advertisements for consumers (Simmons, 2016; Wall, 2018; Wigan & Clarke, 2013). Big data and data analytics in criminology and criminal justice, on the other hand, have seen comparatively limited application (see DeLisi, 2018), primarily in terms of surveillance (e.g., Miller, 2014), predicting the likelihood of recidivism (e.g.. Wall, 2018), and in predictive policing applications (e.g., Brayne, 2017). Furthermore, the application of big data processes and analytics in criminal justice contexts is far from uniform; it varies across different legal jurisdictions, with some criminal justice organisations increasingly making use of big data and data analytics, and others more reluctant to do so, whether in law enforcement, the courts, or corrections (Koper, 2014). Indeed, the challenge of facilitating cultural shifts in organisational thinking in the criminal justice context is far from new and historically has more often than not been met with substantial resistance. Nonetheless. the proliferation of big data has led governments, alongside various components of the criminal justice system, to become increasingly interested in its potential applications.

In the criminal justice context, the appeal of big data reflects the culmination of relatively recent advances in technology, specifically in terms of the ubiquitous distribution of interconnected digital devices (Smith, Bennett Moses, & Chan, 2017), and the increasingly common practice of data-linkage. In terms of the former, a consequence of this interconnectivity is the constant generation and circulation of massive amounts of real-time data, which can range from locational data to metadata lurked to online transactions (Kitchin, 2014). Data-linkage practices are becoming more common. They involve linking large administrative data sets from different government organisations and agencies with the key goal of better informing policy. Big data in criminology and criminal justice therefore primarily encapsulates both the massive data sets generated by interconnected devices and data-linkage, and the analytic techniques and tools that are employed to make sense of such large amounts of data (Chan, 2017). These characteristics set big

Big data in criminology! and criminal justice 9 data and the associated analytics apart from more traditional methods of data collection and statistical analysis in criminology, such as cross-sectional studies with specific samples and the use of survey instruments (Lynch, 2018). In terms of analysis, typical database software and analytic tools commonly used in criminology are increasingly becoming unable to capture, store, manage, and analyse the sheer volume of data and velocity with which it is generated. In other words, big data represents a new frontier for criminology and criminal justice, and in order to optimise the potential of generating positive outcomes and mitigating associated risks it is necessary to become forward thinking when it conies to adapting the tools, practices, and procedures of the trade accordingly, which is the focus of the cunent volume.

Despite its relatively recent appearance on the criminology and criminal justice landscapes, a healthy and robust debate about the implications of big data is emerging globally. To be sure, this is a positive development. At the same time, much of this is speculative and while interesting and thought-provoking, it is additionally necessary to advance the debate by generating empirical evidence about the application of big data and data analytics in criminology and criminal justice. For example, can big data and data analytics be used to determine if law enforcement patterns in a given community reflect over policing of certain sub-groups of the population? Skillicorn, Leuprecht, and Green investigate this very example in the case study of big data analytics and policing that they provide in Chapter 5. Some argue that the use of big data in criminology would shift analytic strategies from making inferences about populations based on specific samples, to analysing data encompassing the entire population (Bennett Moses & Chan, 2014; Chan & Bennett Moses, 2016). In the applied context of criminal justice, a key concern is that decision-making would shift away from individuals with ‘expertise’ and experience to data-driven decisions based on empirical patterns emerging out of big data (Bennett Moses & Chan, 2014; Simmons, 2016). Indeed, these concerns are not new in principle, one can look to the widespread use of, and controversies surrounding, actuarial risk assessment instruments in the context of collections or sentencing grids to guide judicial decisions in different US states to name only a couple. Furthermore, what has emerged from the evidence along these lines is also not surprising; relying exclusively on either a large amount of data without any consideration of context, or vice versa, generalising from single and unique experiences to broader contexts does not typically produce accurate results or optimal outcomes in either scenario. These are some of the first ‘fallacies’ taught in undergraduate research methods courses; the ecological and exception fallacies. Nonetheless, given that the emergence of big data has advanced a shift toward a predictive approach to both knowledge and investigation in criminology and criminal justice (Mantelero & Vaciago, 2015), it seems that scholars with high quality research training are now more relevant than ever. In effect, emerging scholars trained in criminology might increasingly find themselves as ‘culture catalysts’ and/or ‘analytics translators’ in the era of big data and analytics when it comes to criminology and criminal justice.