July 2015
Volume 57, No. 3 | Go to abstracts
Articles
Page 293
Une étude psychométrique des items du Level of Service / Case Management Inventory (LS / CMI) avec la contribution de la théorie classique des tests chez les personnes contrevenantes du Québec
Guy Giguère, Denis Savard et Franca Cortoni
Page 330
Research Access Barriers as Reputational Risk Management: A Case Study of Censorship in Corrections
Tara Marie Watson
Page 363
Situational Action Theory: Cross-Sectional and Cross-Lagged Tests of Its Core Propositions
Gerben J.N. Bruinsma, Lieven J.R. Pauwels, Frank M. Weerman, and Wim Bernasco
Page 399
Penser l’intervention correctionnelle autrement : réflexions critiques sur la prise en charge des justiciables
Manon Jendly, Bastien Quirion, Marion Vacheret et Denis Lafortune
Commentary
Page 424
Simpson’s Paradox in Canadian Police Clearance Rates
Simon Demers and D. Kim Rossmo
Abstracts
Une étude psychométrique des items du Level of Service / Case Management Inventory (LS/CMI) avec la contribution de la théorie classique des tests chez les personnes contrevenantes du Québec
Guy Giguère, Denis Savard, Franca Cortoni
The purpose of an actuarial instrument is to generate a score or numeric index that can be used to predict the probability of an event. The Level of Service / Case Management Inventory (LS/CMI), which is an inventory of the level of service and the management of cases, is part of the last generation of risk-assessment actuarial tools. The predictive validity studies that have been conducted on tools such as the LS/CMI and the Level of Service Inventory-Revised (LSI-R) could not determine satisfactorily the metrological properties of the items that comprise them. The analysis in the present article fills a gap in the writings of specialists by exploiting scientific advancements in measurement theory to the benefit of criminology. The sample was drawn from evaluations carried out at two different times, February 2007 or March 2008, of two groups: 2,670 inmates and 1,837 people serving sentences in the community. The assessments focused primarily on the 43 items operationalizing the eight components of risk that comprise the structure of Part 1 of the LS/CMI. The results of these assessments have helped target some items that are problematic when it comes to discrimination and redundancy.
_____________________________________________
Research Access Barriers as Reputational Risk Management: A Case Study of Censorship in Corrections
Tara Marie Watson
Many criminal justice organizations, including the Correctional Service of Canada (CSC), grant research access through their own research branches. I attempted to interview CSC employees for research about programming and policy in relation to in-prison substance abuse, but access was denied. I have turned my experience into a case study, where I treat my correspondence with CSC as a unique source of data. Although access-denied case studies have appeared in the literature on conducting prison research, I apply a novel lens, reputational risk management, to expand the conceptual toolkit for future researchers. I also use interview data from a sample of 16 participants – former CSC senior administrative officials, former CSC front line staff, and external stakeholders – to supplement my analysis. The case study and interviews reveal new insights regarding access barriers, censorship, and the insular character of CSC research. These restrictions can lead to adverse consequences such as the (re)production of limited knowledge about corrections and the curtailment of innovative solutions to problems. I thus encourage researchers to further refine the application of reputational risk to criminal justice settings and to be persistent in their efforts to access correctional organizations.
_____________________________________________
Situational Action Theory: Cross-Sectional and Cross-Lagged Tests of Its Core Propositions
Gerben J.N. Bruinsma, Lieven J.R. Pauwels, Frank M. Weerman, Wim Bernasco
Situational Action Theory (SAT) is a recently developed general action theory of crime that integrates and synthesizes existing individual and ecological explanations. SAT explicitly states that the individual’s propensity for criminal behaviour (morality and self-control) and exposure to criminogenic settings (rule breaking peers and time spent in unsupervised, unstructured activities) interact to determine whether a crime is committed. In the present article, core assumptions of SAT are tested by estimating cross-sectional and lagged models on two-wave panel data from adolescents in The Hague (The Netherlands). Generally, the findings support SAT, including the situational interaction between morality and self-control. However, the findings also raise questions about SAT. In particular, we did not find lagged effects of morality on later offending, and we found only a few significant interaction effects on offending between the two peer variables and morality and self-control. Generally, there was not much support for the SAT theory that adolescents with low morality or low self-control are more vulnerable to (situational) peer influences. The article concludes with a discussion of how additional situational peer variables may be included in SAT.
_____________________________________________
Penser l’intervention correctionnelle autrement : réflexions critiques sur la prise en charge des justiciables
Manon Jendly, Bastien Quirion, Marion Vacheret, Denis Lafortune
The intervention model developed by the Correctional Service of Canada is an inspiration to many countries around the world. This Canadian model was discussed within focus groups organized in Belgium, France, and Switzerland that offered creative suggestions for the analysis of corrections. The present article presents the main results of our research project, which is about debating ideas around alternative ways to consider correctional intervention. Participants in the focus groups defined correctional intervention as a particular form of encounter that takes place within an institutional setting that is strongly normative. The issues around (1) the context of the intervention, (2) the prisoner, and (3) the nature of the encounter have been explored in order to provide recommendations about the renewal of the way to consider intervention within correctional facilities.
_____________________________________________
Simpson’s Paradox in Canadian Police Clearance Rates
Simon Demers, D. Kim Rossmo
Police clearance rates and other forms of aggregated criminal justice data can be susceptible to statistical artefacts such as Simpson’s paradox (Yule-Simpson effect). Simpson’s paradox occurs when a trend apparent in separate data groups reverses itself once the groups are combined. Within 15 years of clearance rate data from the 50 largest Canadian police jurisdictions, 210 instances of Simpson’s paradox were discovered (annual mean = 14.0). These instances included four cases in which a reversal occurred simultaneously in all crime categories and subcategories. This finding suggests the need for caution when using clearance rates as a comparative measure of police performance, particularly between jurisdictions or time periods with different crime mixes. Criminal justice researchers, policy makers, and crime analysts should be aware of Simpson’s paradox and its potential effect on aggregated data. Finally, the possibility of a double reversal must be considered when attempting to resolve Simpson’s paradox.