Leslie Humphreys

A Mixture Model for Longitudinal Partially Ranked Data

This article discusses the use of mixture models in the analysis of longitudinal partially ranked data, where respondents, for example, choose only the preferred and second preferred out of a set of items. To model such data we convert it to a set of paired comparisons. Covariates can be incorporated into the model. We use a nonparametric mixture to account for unmeasured variability in individuals over time. The resulting multi-valued mass points can be interpreted as latent classes of the items. The work is illustrated by two questions on (post)materialism in three sweeps of the British Household Panel Survey.

Using the UK general offender database as a means to measure and analyse organised crime

Purpose – Organised Crime is notoriously difficult to identify and measure, resulting in limited empirical evidence to inform policy makers and practitioners. The purpose of this paper is to explore the feasibility of identifying a greater number of organised crime offenders, currently captured but invisible, within existing national general crime databases. Design/methodology/approach – All 2.1 million recorded offenders, captured over a four-year period on the UK Police National Computer, were filtered across three criteria associated with organised crime (co-offending, commission of specific offences, three years imprisonment or more). The 4,109 “organized crime” offenders, identified by the process, were compared with “general” and “serious” offender control groups across a variety of personal and demographic variables. Findings – Organised crime prosecutions are not random but concentrate in specific geographic areas and constitute 0.2 per cent of the offender population. Offenders can be differentia...

Older People and Dissatisfaction with Wheelchair Services

Britain has experienced an unprecedented increase in wheelchair use during the past two decades. The authors take a social model approach to disability and report on their study in the north-west of England into the social implications of this increase. Qualitative interviews and a large-scale social survey reveal the circumstances of users and their experiences. Building on their descriptive statistics, the authors use latent class analysis to identify from amongst their respondents the characteristics of dissatisfied users of UK National Health Service wheelchair provision: they were more likely to be older, frailer females; and to be living in residential and nursing care homes. Prescribing practices and policy guidelines are analysed to consider how a hierarchy of need is operating to determine wheelchair allocation.

Middle-Class Offenders as Employees—Assessing the Risk: A 35-Year Follow-Up

A 35-year follow-up of a series of 317 middle-class offenders in England and Wales suggests that the dangers of employing offenders may be more limited than expected. Although 40% were subsequently convicted, only 8% were subsequently convicted of offenses that directly and adversely affected an employer. This work should challenge the “exaggerated fears” of employers. Interestingly, variables which normally predict subsequent criminal activity made no impact in trying to predict offenses against an employer.

Understanding Deception: Disentangling Skills from Conviction

Deception is often associated with economic gain and white-collar crime. But studying deception highlights the need for criminologists and practitioners to move beyond legal definitions and conviction rates when attempting to achieve depth in understanding criminality, its motivations and possible specialisms. Further, to explore the complexity of deception requires recognition of the range of skills inherent in this modus operandi, which is better recognised as a potentially-criminal tool found in much criminal behaviour. Theories that attempt to explain specialisation need to move on from a focus on crimes committed and give appropriate attention to skills employed.