Against Marginality: Arguments for a Public Criminology

Despite its accumulated theoretical and empirical heft, the discipline of criminology has had distressingly little impact on the course of public policy toward crime and criminal justice. This article addresses the sources of that troubling marginality, with special emphasis on the powerful disincentives to greater public impact that operate within the discipline itself and the research universities that mainly house it—including the pressure to publish ever more narrow research in peer-reviewed journals at the expense of efforts at synthesis and dissemination that could serve to educate a broader public. Achieving a greater voice in the world outside the discipline will require a concerted move toward a more explicitly public criminology, and seeing to it that the work of such a criminology is more reliably supported and rewarded within the universities and the profession as a whole.

Which Women are More Likely to be Abused? Public Housing, Cohabitation and Separated/Divorced Women.

Data collected at Canadian public housing estates in eastern Ontario are used here to analyze two hypotheses. Overall these women report more violence than do otherwise situated women in other general surveys. More specifically, complex theoretical models were designed to generate two hypotheses for further analysis: First, that separated/divorced women are more likely to be abused within public housing than married women. Second, that cohabiting women will report violence victimization at a higher rate than separated, divorced, or married women. Some support for both hypotheses were found, and the theoretical models are used to discuss these findings.

Court Based Psychiatric Assessment: A Case for an Integrated and Diversionary Public Health Role

This article describes a follow-up study of 232 individuals who underwent psychiatric assessment by a Criminal Justice Mental Health Team (CJMHT) in 2001/2002, and also draws upon in-depth interviews conducted with 26 of the cohort. At assessment many people are identified with substance misuse problems, as homeless and with a history of psychiatric contact but in the main their problems are of insufficient severity to merit diversion to psychiatric hospital. The study mapped service contact, housing and offending in the 12 months following assessment and compared this to the 12 months prior to assessment, and found increased levels of service contact but also increased levels of offending and no decrease in homelessness. Thus assessment by the CJMHT brought few discernible advantages for the majority of clients. This was also the perception of the 26 clients who were interviewed. Their own perceptions of their lifestyle and the support that they deemed most valuable are described to identify means of enhancing the efficacy of court assessment.

A comparison of methods for classifying clinical samples based on proteomics data : a case study for statistical and machine learning approaches

The discovery of protein variation is an important strategy in disease diagnosis within the biological sciences. The current benchmark for elucidating information from multiple biological variables is the so called “omics” disciplines of the biological sciences. Such variability is uncovered by implementation of multivariable data mining techniques which come under two primary categories, machine learning strategies and statistical based approaches. Typically proteomic studies can produce hundreds or thousands of variables, p, per observation, n, depending on the analytical platform or method employed to generate the data. Many classification methods are limited by an n≪p constraint, and as such, require pre-treatment to reduce the dimensionality prior to classification. Recently machine learning techniques have gained popularity in the field for their ability to successfully classify unknown samples. One limitation of such methods is the lack of a functional model allowing meaningful interpretation of results in terms of the features used for classification. This is a problem that might be solved using a statistical model-based approach where not only is the importance of the individual protein explicit, they are combined into a readily interpretable classification rule without relying on a black box approach. Here we incorporate statistical dimension reduction techniques Partial Least Squares (PLS) and Principal Components Analysis (PCA) followed by both statistical and machine learning classification methods, and compared them to a popular machine learning technique, Support Vector Machines (SVM). Both PLS and SVM demonstrate strong utility for proteomic classification problems.