Evaluating and comparing profiles of burglaries developed using three statistical classification techniques: cluster analysis, multidimensional scaling, and latent class analysis

Fox, Bryanna; Escue, Melanie

doi:10.1080/1068316x.2021.1880582

Cited by 7 publications

(5 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The protoscript was adapted by using the existing protoscript framework described in Leclerc et al (2011) and Boxall et al (2018) and adding stages to better highlight the sequence of events evident across incidents (e.g., victim response and offender reaction to interventions). Once the protoscript was developed, further analysis was conducted to explore diversity across the sample using cluster analysis in SPSS (v. 27.0) (Fox & Escue, 2021 ). In total, 38 variables were analyzed and recorded as present or absent (1,0).…”

Section: Methodsmentioning

confidence: 99%

Different Scripts, Different Casts: A Crime Script Analysis Indicating Intimate Partner Violence Is Not All the Same

Carney

Kebbell

Eriksson

et al. 2023

Violence Against Women

View full text Add to dashboard Cite

Crime script analysis was used to analyze intimate partner violence diversity by identifying variables significantly associated with different script tracks. Qualitative thematic analysis using official police administrative data from Queensland, Australia, was used to develop an intimate partner violence protoscript ( n = 40), followed by quantitative hierarchical cluster analysis and cross-tabulations to examine diversity within scripts. Four diverse script tracks were identified: “escalating jealousy,” “persistently possessive,” “controlling victim agency,” and “enduring argument.” Intimate partner diversity exists with divisions based on statistically significant variables. Implications for situational crime prevention and the use of mixed methods for strengthening crime script analysis are discussed.

show abstract

Section: Methodsmentioning

confidence: 99%

Different Scripts, Different Casts: A Crime Script Analysis Indicating Intimate Partner Violence Is Not All the Same

Carney

Kebbell

Eriksson

et al. 2023

Violence Against Women

View full text Add to dashboard Cite

show abstract

“…Second, to determine heterogeneity among missing person case characteristics, MDS was used with all dichotomous variables. The use of MDS is appropriate to identify the structure in a set of distance measures between a single set of objects or cases and have been previously used in criminological studies (e.g., Fox & Escue, 2021; García‐Barceló et al., 2020; Lehmann et al., 2013; Sea & Beauregard, 2021). Specifically, observations are assigned to specific locations in a conceptual low‐dimensional space so that the distances between points in the space match the given similarities and dissimilarities as closely as possible (Giguère, 2006; Jaworska & Chupetlovska‐Anastasova, 2009; Tsogo et al., 2000).…”

Section: Methodsmentioning

confidence: 99%

An exploration of unsolved missing persons cases suspected of a criminal outcome: A forensic victimology approach

Chopin

Beauregard

Baroche³

et al. 2022

Journal Invest Psychology

View full text Add to dashboard Cite

The purpose of this study is to explore cases of missing persons suspected of being criminal. Specifically, this research aims to empirically describe the circumstances surrounding criminal disappearances and examine whether there are different subcategories within these cases using a victimological framework. The data used in this study come from an operational police database. The sample includes 155 unsolved missing person cases whose thorough investigation by the police concluded that a criminal motive remained the most likely hypothesis. These cases occurred on the entire territory of metropolitan France and overseas. Multidimensional scale analysis was used to explore the context of disappearance according to missing person characteristics, lifestyle, everyday activities, and situational exposure at the time of the disappearance. Results show that the distribution of variables in a two-dimensional space reveals four distinct different categories related to lifestyle and situational exposure: riskier lifestyle, riskier situation, safer situation, and safer lifestyle. This study is the first to empirically explore unsolved missing person cases suspected of a criminal outcome. Moreover, the framework used in this study demonstrates the importance of victimology in an investigative context.

show abstract

“…First, multiple goodness‐of‐fit measures are used to accurately select the best number of classes for interpretation: the Akaike information criterion (AIC), the Bayesian information criterion (BIC), and the consistent Akaike information criterion (CAIC). By comparison, clinical methods and classification techniques such as multidimensional scaling have more subjective criteria for determining the number and nature of resultant profiles (Fox & Escue, 2021). Second, LCA allows for multiple categories in a variable, whereas related techniques such as cluster analysis can only be utilized with dichotomous variables, limiting important heterogeneity in measurement.…”

Section: Methodsmentioning

confidence: 99%

“…Second, LCA allows for multiple categories in a variable, whereas related techniques such as cluster analysis can only be utilized with dichotomous variables, limiting important heterogeneity in measurement. Finally, LCA is better suited for examining samples of offenders as it does not assume that the data are normally distributed and linear, as these are common issues for data used in criminological research (Fox & Escue, 2021).…”

Section: Methodsmentioning

confidence: 99%

Juveniles arrested for murder: A latent class analysis of male offenders

Fox

Heide

Khachatryan

et al. 2021

Behavioral Sci & The Law

Self Cite

View full text Add to dashboard Cite

Researchers and practitioners have a need for valid and generalizable typologies of juvenile homicide offenders (JHOs) to better understand the heterogeneous nature of JHOs, and use this knowledge to inform prevention efforts. Prior studies of JHOs have typically utilized a clinical approach, which is rich in data but based on small nonrepresentative samples, or relied on larger aggregate datasets which are more reliable but have few measures available. This study aimed to address these limitations by examining the latent heterogeneity (i.e. unapparent differences) in JHOs, their victims, and incident characteristics among more than 40,000 male JHOs arrested in the United States between 1976 and 2016 using latent class analysis, to allow for the objective identification of latent taxonomies. Six distinct subtypes of male JHOs are found in the analysis, which contributes to the theoretical and substantive knowledge base regarding JHOs, and may be used to develop and implement more effective interventions and violence prevention programs for these young offenders.

show abstract

Evaluating and comparing profiles of burglaries developed using three statistical classification techniques: cluster analysis, multidimensional scaling, and latent class analysis

Cited by 7 publications

References 53 publications

Different Scripts, Different Casts: A Crime Script Analysis Indicating Intimate Partner Violence Is Not All the Same

Different Scripts, Different Casts: A Crime Script Analysis Indicating Intimate Partner Violence Is Not All the Same

An exploration of unsolved missing persons cases suspected of a criminal outcome: A forensic victimology approach

Juveniles arrested for murder: A latent class analysis of male offenders

Contact Info

Product

Resources

About