Identifying Important Factors to Prevent Juvenile Delinquency among Male and Female Adolescents: an Exploratory Analysis Using the LASSO Regression Algorithm in the Korean Children and Youth Panel Survey (KCYPS)

Choi, Jungtae

doi:10.1007/s12187-022-09916-6

Cited by 3 publications

(3 citation statements)

References 77 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 1 presents an overview of three studies predicting criminal behavior and two studies predicting substance use. The most important predictors found in the studies correspond to the previously presented literature, including social factors such as friends' substance use (Vázquez et al, 2020) or peer and romantic relationships (Choi, 2022), individual factors such as impulsivity (Afzali et al, 2018;Pelham et al, 2020) or aggression (Choi, 2022) as well as demographic factors such as gender (Vázquez et al, 2020) and race (Neuilly et al, 2011). Methodologically, the findings are inconclusive regarding the additional benefit of more complex modeling techniques over simpler models: ML models outperformed simpler models in some studies (Afzali et al, 2018;Neuilly et al, 2011), but not all (Pelham et al, 2020), whereas the others did not even include linear models as baseline.…”

Section: Machine Learning Algorithms In the Prediction Of Deviant Beh...supporting

confidence: 84%

Predicting Juvenile Delinquency and Criminal Behavior in Adulthood Using Machine Learning

Schroeders¹,

Mariss²,

Jankowsky³

2023

Preprint

View full text Add to dashboard Cite

By violating social norms, deviant behavior is an important issue that affects society as a whole and also has serious consequences for its individuals. Different scientific disciplines have proposed theories on deviant behavior, which often fall short of predicting actual behavior. In this registered report, we will use data of the longitudinal National Study of Adolescent to Adult Health (Add Health) to examine the predictability of juvenile delinquency and criminal behavior in adults, differentiating between drug-related, property, and violent offenses. We propose to compare the predictive accuracy of traditional regression models to different machine learning algorithms (elastic net regressions and gradient boosting machines). Furthermore, we will explore the differences in the selection of the most important predictors of juvenile delinquency vs. adult criminal behavior to consider developmental changes and potentially identify risk factors at an early stage.

show abstract

Section: Machine Learning Algorithms In the Prediction Of Deviant Beh...supporting

confidence: 84%

Predicting Juvenile Delinquency and Criminal Behavior in Adulthood Using Machine Learning

Schroeders¹,

Mariss²,

Jankowsky³

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Adolescence shows stronger stability for depression and delinquency than other life stages. Moreover, adolescent depression and delinquency are symptoms that manifest continuously and are associated with more serious problems later in adulthood (81)(82)(83). Therefore, early detection and intervention are important.…”

Section: Discussionmentioning

confidence: 99%

Longitudinal relationship between depression and antisocial behaviors in Korean adolescents

Kim

Cho

et al. 2023

Front. Psychiatry

View full text Add to dashboard Cite

BackgroundIt is well known that depression and delinquency in adolescents are highly correlated, but longitudinal studies on the causal relationship between them are not active in East Asia compared to in Western culture. In addition, even the results of research on causal models and sex differences are inconsistent.ObjectivesThis study examines the longitudinal reciprocal effects between depression and delinquent behavior in Korean adolescents based on sex differences.MethodsWe conducted multiple-group analysis by using an autoregressive cross-lagged model (ACLM). Longitudinal data from 2,075 individuals (2011–2013) were used for analysis. The longitudinal data are from the Korean Children and Youth Panel Survey (KCYPS), and data were used beginning with students at 14 years old (in the second grade of middle school) and tracked them until they were 16 (in the first grade of high school).ResultsBoys’ delinquent behaviors at 15 years (the third grade of middle school) affected their depression at 16 years (the first grade of high school). In contrast, girls’ depression at 15 years (the third grade of middle school) influenced their delinquent behaviors at 16 years (the first grade of high school).DiscussionThe findings support the failure model (FM) among adolescent boys and the acting-out model (ACM) among girls. The results imply that strategies to effectively prevent and treat delinquency and depression in adolescents must consider sex effects.

show abstract

“…Deep learning represents both fascinating and terrifying potentialities as a type of AI that can automatically adapt with minimal to no human interference. This method with various algorithms has already been used to predict things like alcohol and substance abuse (Afzali et al, 2019; Vázquez et al, 2020), delinquent behavior (Choi, 2022), and general recidivism (Neuilly et al, 2011). Cockerill (2020) warned that forensic evaluators will be woefully unprepared if they fail to carefully consider the impact of AI and deep learning on forensic evaluations in general and violent risk assessment specifically.…”

Section: Emerging Advancementsmentioning

confidence: 99%

Child and adolescent forensic evaluations: Challenges amid scientific and technological advancements.

Kennedy,

Woods,

Anillo

et al. 2023

Practice Innovations

View full text Add to dashboard Cite

Identifying Important Factors to Prevent Juvenile Delinquency among Male and Female Adolescents: an Exploratory Analysis Using the LASSO Regression Algorithm in the Korean Children and Youth Panel Survey (KCYPS)

Cited by 3 publications

References 77 publications

Predicting Juvenile Delinquency and Criminal Behavior in Adulthood Using Machine Learning

Predicting Juvenile Delinquency and Criminal Behavior in Adulthood Using Machine Learning

Longitudinal relationship between depression and antisocial behaviors in Korean adolescents

Child and adolescent forensic evaluations: Challenges amid scientific and technological advancements.

Contact Info

Product

Resources

About