Modelling escalation in crime seriousness: a latent variable approach

Abstract

This paper investigates the use of latent variable models in assessing escalation in crime seriousness. It has two aims. The first is to contrast a mixed-effects approach to modelling crime escalation with a latent variable approach. The paper therefore examines whether there are specific subgroups of offenders with distinct seriousness trajectory shapes. The second is methodological-to compare mixed-effects modelling used in previous work on escalation with group-based trajectory modelling and growth mixture modelling (mixture of mixed-effects models). The availability of software is an issue, and comparisons of fit across software packages is not straightforward. We suggest that mixture models are necessary in modelling crime seriousness, that growth mixture models rather than group-based trajectory models provide the best fit to the data, and that R gives the best software environment for comparing models. Substantively, we identify three latent groups, with the largest group showing crime seriousness increases with criminal justice experience (measured through number of conviction occasions) and decreases with increasing age. The other two groups show more dramatic non-linear effects with age, and non-significant effects of criminal justice experience. Policy considerations of these results are briefly discussed.

Trajectories of resilience to acute malnutrition in the Kenyan drylands

“…We use a latent class mixed model (LCMM) for the classification of ward-level nutrition dynamics into distinct resilience trajectories (Proust-Lima et al, 2014, 2016, 2022Sène et al, 2014;Andrinopoulou et al, 2021). Latent class models have successfully been applied in the fields of criminology (Francis and Liu, 2015;Stone et al, 2023) and psychology (Simon et al, 2022), and more recently in human epidemiology (Lennon et al, 2018). We follow the eight-step framework proposed by Lennon et al (2018) and the Guidelines for Reporting on Latent Trajectory Studies ("GRoLTS-Checklist") (Van De Schoot et al, 2017) (see Supplementary Table S2) for model development and implement the analysis in R, using the LCMM package (Proust-Lima and Liquet, 2011, 2016) (see Supplementary Section 2.1).…”

The Use of a Hidden Mixture Transition Distribution Model in Clustering Few but Long Continuous Sequences: An Illustration with Cognitive Skills Data

“…The Growth Mixture Model (GMM) model has become a reference in the continuous longitudinal data modeling, with various applications in criminology [13], health and medicine [14,15], psychology and social sciences [16][17][18], among others (see [19]). The GMM [14,20] is a model designed to discover and describe the unknown groups of sequences that share a similar pattern.…”

“…This method may be represented as a mixture of mixed-effects models, where each of the unknown sub-populations follows a distinct linear mixed effect model. Its main advantage over previous similar models is that it allows for the estimation of a specific variance-covariance structure within each class [13]. Within-class inter-individual variation is allowed for the latent variables via distinct intercept and slope variances, which are represented by a class-specific fixed effects and random effects distribution.…”

Markovian-Based Clustering of Internet Addiction Trajectories

“…This method may be represented as a mixture of mixed-effects models in which each of the unknown subpopulations follows a distinct linear mixed-effects model. Its main advantage over other similar models-like the heterogeneity model (Verbeke and Lesaffre 1996)-is that it allows for estimation of a specific variance-covariance structure within each class (Francis and Liu 2015). Within-class inter-individual variation is possible for latent variables via distinct intercept and slope variances, represented by a class-specific fixed-effects and random-effects distribution.…”