Environmental criminology in the big data era

Abstract

This study examines to what extent new and emerging data sources or big data have been empirically used to measure key theoretical concepts within environmental criminology. By means of a scoping review, aimed at studies published between 2005 and 2018, insight is provided into the characteristics of studies that used big data sources within environmental criminology. The type and extent of big data sources used, as well as the strengths and weaknesses of these data sources, are synthesized. After the selection procedure, 84 studies were included for further analysis. Although the number of studies increased each year, there has been a remarkable increase in the number of studies since 2014. The findings suggest that most studies used administrative data or user-generated content as one type of research data. However, innovative data sources (automated and volunteered data) have gained in importance in recent years. Also, most studies are of a descriptive or predictive nature, predominantly conducted by computational (social) scientists. Since these approaches pay little to no attention to mechanisms that bring about social outcomes, an alternative philosophical framework is proposed. We put forward a scientific realist approach as a solution to integrate data-driven and theory-driven research. This approach responds to recent calls to move towards an ‘analytical criminology’. The results are discussed within this framework, and translated into avenues for future research.

Jumpstarting the Justice Disciplines: A Computational-Qualitative Approach to Collecting and Analyzing Text and Image Data in Criminology and Criminal Justice Studies

“…Computational techniques for collecting and analyzing text and image data are becoming increasingly popular in criminology and criminal justice studies (Al-Zaidy et al 2012;Khorshidi et al 2021;Kleck et al 2006;Kuang et al 2017;Pina-Sánchez et al 2019b,a;Snaphaan and Hardyns 2019;Vander Beken et al 2021;Williams and Burnap 2016). Although much of the discussion and application of computational methods to date pertains to quantitative research, there is much to be gained from the advancement and application of computational techniques in support of qualitative research too.…”

“…Reviewing this documentation ahead of time is also important for algorithmic design, as many websites will ask, for example, that users moderate the speed of their requests in order not to delay server access by others. Such data, however, are all "readymades", which differs from "custommades" (Salganik 2019, 7), sometimes also referred to as "found data" and "made data" (Connelly et al 2016;Snaphaan and Hardyns 2019). Whereas web scraping can be an effective means of collecting and repurposing readymade data available online (i.e., data not created for the purposes of research), sometimes the research question we are trying to answer warrants the collection of custommades (i.e., data that we explicitly create for the purposes of our research).…”

Do Mobile Phone Data Provide a Better Denominator in Crime Rates and Improve Spatiotemporal Predictions of Crime?

“…[36] (p. 263). These new and emerging data sources, such as GPS data and mobile phone data, provide large opportunities for research in environmental criminology [2,15].…”

“…In many fields, decision-making processes are increasingly based on intelligence gained from big data, complex datasets containing large amounts of data, from which new information can be extracted. Although the use of big data is relatively new in criminology, there are a lot of opportunities to increase our knowledge and improve data-based applications by leveraging big data [1,2]. This is particularly true for intelligence-led policing, with its focus on data-based, proactive policing [3].…”

Big Data, Data Mining, Machine Learning, and Deep Learning Concepts in Crime Data

“…The identification and categorization of these behaviours imply an understanding of the emotional state of the suspect himself/herself and of the society which includes him/her (Umair, et al, 2015). For this reason, crime statistics are important, and models which include a range of dynamic micro or macro areas that predict crime behaviour with hypothetical and experimental crime models have been developed in order to understand crime (Snaphaan & Hardyns, 2019;Bulgakova et al, 2019). However, the main problem in crime data is that it is in the form of a complex and large data stack before analysis.…”