Analyzing the offending activity of inmates: Trajectories of offense seriousness, escalation, and de-escalation. (May 2017)

• Record Type: Journal Article
• Authors: Cihan, Abdullah; Sorensen, Jonathan; Chism, Kimberly A.
• Is Part Of: Journal of criminal justice. Number 50(2017)
• Page Start: 12
• Page End: 18
• Publication Date: 2017-05
• Subjects: Criminal career paradigm -- De-escalation -- Escalation -- Group-based trajectory modeling -- Inmate misconduct -- Multinomial logistic regression; Criminal justice, Administration of -- Periodicals; Justice pénale -- Administration -- Périodiques; 364.05
• DOI: 10.1016/j.jcrimjus.2017.04.001
• Languages: English
• ISSNs: 0047-2352

Abstract

Abstract: Purpose: The current study seeks to systematically identify developmental patterns in the seriousness (escalation and de-escalation) of inmate misconduct and the factors that distinguish those patterns. Methods: Official data from a northwestern state prison system and criminal history files dating between December 1996 and December 2008 yielded an excess of 30, 000 inmates. Inmates who served a minimum of three years of continuous custody were included, resulting in a final sample of 8351 inmates. Results: A group-based trajectory analysis identified three distinct groups of inmates based on their developmental trajectories of offense seriousness: escalators, de-escalators, and statics. A multinomial logistic regression revealed that the rate of prior arrests was the most consistent predictor for each group. Only membership in the static group of misconduct violators was significantly associated with a broad range of predictors, such as age, gender, and custody level. This suggests that variables found to be consistently related prison misconduct in the literature may be not be as successful at predicting other developmental patterns of inmate misconduct. Conclusions: The trajectory analysis indicates that the inmate population is not homogenous and subgroups of inmates with unique trajectories of misconduct seriousness exist in the inmate population. Highlights: Three inmate trajectories were identified: escalators, de-escalators, and statics. Rate of prior arrests was the most consistent predictor for each group. Race, ethnicity, and sexual offenses predicted static and escalator groups. Custody level was significantly associated with statics. Rate of violent arrests and rate of prior arrests predicted de-escalators.