Patterns of Problem Behaviors and Predictors of Class Membership among Adolescents in the Republic of Korea: A Latent Class Analysis

Abstract

Although subgroups of adolescent problem behaviors (PBs) may exist and have different characteristics, most available studies have focused on exploring a single PB. Thus, we aimed to investigate latent classes of adolescent PBs and to identify important predictors of latent class membership. We analyzed nationally representative secondary data—Waves 4 and 5 of the Korean Children & Youth Panel Survey—obtained from the 2010 cohort of seventh graders and their parents. Specifically, using seven PBs (e.g., daily smoking, monthly drinking, and sexual intercourse) measured in Wave 5, we conducted a latent class analysis (LCA) to identify the model that best fit the data. In the next step, we conducted an LCA with covariates to investigate Wave-4 predictors of latent class membership. In our study, a three-latent-class model best fit the data: the Low Risk class (78%) characterized by low probabilities of engagement in all PBs, Non-Habitual Alcohol Use class (14%), and Habitual Cigarette and Alcohol Use class (7%). In addition, successful predictors of latent class membership included gender, parental education, friendships, relationships with teachers, parental affection, abuse inflicted by parents, and aggression. Health professionals should develop interventions tailored to each homogeneous subgroups of PBs in order to obtain more effective outcomes. Additionally, when developing these interventions, they should consider multilevel characteristics (e.g., individual, peer, and parental factors) that differentiate these subgroups.