• The Korean Journal of Applied Statistics
• /
• v.30 no.5
• /
• pp.615-632
• /
• 2017

Unlike randomized trial, statistical strategies for inferring the unbiased causal relationship are required in the observational studies. The matching with the propensity score is one of the most popular methods to control the confounders in order to evaluate the effect of the treatment on the outcome variable. Recently, new methods for the causal inference in latent class analysis (LCA) have been proposed to estimate the average causal effect (ACE) of the treatment on the latent discrete variable. They have focused on the application study for the real dataset to estimate the ACE in LCA. In practice, however, the true values of the ACE are not known, and it is difficult to evaluate the performance of the estimated the ACE. In this study, we propose a method to generate a synthetic data using the propensity score in the framework of LCA, where treatment and outcome variables are latent. We then propose a new method for estimating the ACE in LCA and evaluate its performance via simulation studies. Furthermore we present an empirical analysis based on data form the 'National Longitudinal Study of Adolescents Health,' where puberty as a latent treatment and substance use as a latent outcome variable.

• Communications of Mathematical Education
• /
• v.34 no.4
• /
• pp.485-506
• /
• 2020

The purpose of this study is to analyze of the effect in Mathematics Teachers beliefs on their students beliefs by Latent Class Regression Model (LCRM). For this analysis, the study used the findings and surveys of Kang, Hong (2020) who developed a belief profile by analyzing the mathematical beliefs of 60 high school teachers and 1,850 second-year high school students learning from them through the Latent Class Analysis (LCA). As a result It was observed that 'Nature of Mathematics', 'Mathematic Teaching' and 'Mathematical Ability' of mathematics teachers beliefs influence the mathematical beliefs of students. The teacher's belief of 'Nature of Mathematics' statistically significant effects on students' beliefs in 'School Mathematics', 'Problem Solving', 'Mathematics Learning'. The teacher's belief of 'Teaching Mathematics', 'Mathematical Ability' statistically significant effects on students' beliefs in 'School Mathematics', 'Problem Solving', 'Self-Concept'. The results of this study can give a preview of the phenomenon in which teacher's mathematical beliefs are reproduced into student's mathematical beliefs. In addition, the results of observation of this study can be used to the contents that can achieve the purpose of reorientation for mathematics teachers.

• The Korean Journal of Applied Statistics
• /
• v.27 no.7
• /
• pp.1077-1095
• /
• 2014

Unlike randomized trial, statistical strategies for inferring the unbiased causal relationship are required in the observational studies. Recently, new methods for the causal inference in the observational studies have been proposed such as the matching with the propensity score or the inverse probability treatment weighting. They have focused on how to control the confounders and how to evaluate the effect of the treatment on the result variable. However, these conventional methods are valid only when the treatment variable is categorical and both of the treatment and the result variables are directly observable. Research on the causal inference can be challenging in part because it may not be possible to directly observe the treatment and/or the result variable. To address this difficulty, we propose a method for estimating the average causal effect when both of the treatment and the result variables are latent. The latent class analysis has been applied to calculate the propensity score for the latent treatment variable in order to estimate the causal effect on the latent result variable. In this work, we investigate the causal effect of adolescents delinquency on their substance use using data from the 'National Longitudinal Study of Adolescent Health'.

• Korean Journal of Childcare and Education
• /
• v.16 no.1
• /
• pp.115-143
• /
• 2020

Objective: The purpose of this study was to examine underlying types of developmental trajectories of adaptation to child care among infants and toddlers. This study also aimed to identify latent classes in their child care adaptation types in order to find predictors that account for individual differences. Methods: Participants were 420 mothers of infants and toddlers and 123 teachers. The levels of child care adaptation of participating infants and toddlers were rated monthly from early April to June, 2019. The collected data were analyzed using growth mixture modeling, latent class analysis and multinominal logistic analysis. Results: The results of growth trajectories of child care adaptation showed there were two to four latent groups by dimension of child care adaptation. Also, the groups of individual dimensions of child care adaptation were classified into three latent classes, which were 'complying and positive group', 'negative group', and 'individualized group. Multinominal logistic analysis revealed that children's age, gender, and temperament differentiated the three latent classes of adaptation to child care. Conclusion/Implications: The results show individual characteristics that infants and toddlers possess should be prudently considered in order for successful adaptation to child care.

• The Journal of the Korea Contents Association
• /
• v.11 no.12
• /
• pp.386-394
• /
• 2011

Recently it is difficult to draw an improvement for customer satisfaction because the ratio of satisfied customers increases in customer satisfaction survey. In addition, the effectiveness of practical application of customer satisfaction survey decreases due to its constitution limitation on its data analysis. In order to solve these problems, it is necessary to develop a novel research to identify the strategy meanings and find dissatisfied factors of satisfied customers using the satisfied customers' reclassification. This study focuses on the satisfied customer segmentation based on Latent Class Analysis (LCA). The case study with high-speed internet service customers show that the satisfied customers are divided into three subgroups using LCA and we draw meaning results such as satisfaction and dissatisfaction factors through analyzing each group. This study is expected to play the role as the groundwork for the revitalization of customer satisfaction survey as well as improving customer satisfaction management.

• 한국노년학
• /
• v.41 no.4
• /
• pp.567-586
• /
• 2021

This study explored what latent classes appear according to the combination of Belief in a Just World (BJW) and social class of middle-aged adults, and examined whether there are differences in career transiton, perceiving a calling, and working as meaning in each class and what characteristics they have. 224 middle-aged people who experienced turnover through online and offline were surveyed and analyzed by Latent Profile Analysis. The participants were divided into 5 latent classes such as; 'Relative self-satisfaction', 'Social contentment', 'Relative deprived', 'Fairness trust' and 'Fairness distrust'. According to the results of MANOVA analysis to figure out if there are differences in career transitions, perceiving a calling, and working as meaning depending on latent classes, significant differences were appeared among latent classes. Finally, multinominal logistic regression analysis was conducted to examine whether demographic characteristics and 'decent work' affect the latent group classification. As a result, the more 'decent work', the higher the probability of belonging to the class with high BJW and social class. On the basis of the results of this study, the implications on the case conceptualization and counseling strategy for adults focusing on BJW and Social class in adults and future research were discussed.

• Journal of Digital Convergence
• /
• v.16 no.6
• /
• pp.385-397
• /
• 2018

This study used latent class analysis to identify heterogeneous subgroups with respect to health condition among adolescents. We also examined associations between latent classes and verified to determine how the patterns of health condition relate to social capital. This study used data from Korean Survey on the Rights of Children and Youth in 2015, which consists of 6,912 from middle and high school students. The findings are as follows. Latent class analysis revealed a three-class solution. Results indicated that family social capital and school capital significantly verified to the above latent classes all family type. But, community social capital not significantly predicted to the above latent classes only single parent families. Policy implications for improving the health condition of adolescents are discussed.

• Communications for Statistical Applications and Methods
• /
• v.23 no.2
• /
• pp.131-146
• /
• 2016

In research on behavioral studies, significant attention has been paid to the stage-sequential process for longitudinal data. Latent class profile analysis (LCPA) is an useful method to study sequential patterns of the behavioral development by the two-step identification process: identifying a small number of latent classes at each measurement occasion and two or more homogeneous subgroups in which individuals exhibit a similar sequence of latent class membership over time. Maximum likelihood (ML) estimates for LCPA are easily obtained by expectation-maximization (EM) algorithm, and Bayesian inference can be implemented via Markov chain Monte Carlo (MCMC). However, unusual properties in the likelihood of LCPA can cause difficulties in ML and Bayesian inference as well as estimation in small samples. This article describes and addresses erratic problems that involve conventional ML and Bayesian estimates for LCPA with small samples. We argue that these problems can be alleviated with a small amount of prior input. This study evaluates the performance of likelihood and MCMC-based estimates with the proposed prior in drawing inference over repeated sampling. Our simulation shows that estimates from the proposed methods perform better than those from the conventional ML and Bayesian method.

• Korean Journal of Social Welfare Studies
• /
• v.40 no.1
• /
• pp.197-234
• /
• 2009

Typological theories of offending postulate that childhood-onset delinquents have a high likelihood of being serious and chronic offenders and that there are a distinct set of risk factors predicting early-onset antisocial behaviors. It is useful to empirically classify children into subgroups based on their deviant behaviors because it helps us to identify unique factors associated with each subgroup. Using the first two waves of the Korean Youth Panel Survey, Elementary School Data, this study aimed(a) to empirically classify 5th graders into latent delinquent subgroups, and (b) to examine the impact of individual, familiar, school, and peer factors on the latent delinquent classes. Latent class analysis yielded three latent classes based on 15 indicators of deviant behaviors - delinquent class, low-level delinquent class, & normative class. The results from multivariate multinomial logistic regression analyses revealed that being male, reporting low self-control, coming from poor family, high association with deviant peers, and being bullied increased the risk of being in the delinquent class. Moreover, low self-control, aggression, domestic violence, low level of attachment to teachers, and deviant peers independently increased the risk of being in the low-level delinquent class compared to the normative class. Based on the study findings, implications for practice as well as future studies were discussed.

• Korean Journal of Health Education and Promotion
• /
• v.32 no.5
• /
• pp.17-31
• /
• 2015

Objectives: This paper aims to identify the health related behaviors patterns and its associated factors among marriage immigrant women in Korea, and discusses their application to health promotion strategies. Methods: The study participants were 7,591 immigrant wives in Gyeonggi province who participated in health examinations conducted by the Korea Association of Health Promotion in 2011-2013. The participants completed self-administered questionnaires on sociodemographics, psychological characteristics, health status and health care factors, and health related behaviors. Results: A 3-latent-class model of health behaviors was identified related to 'lack of physical activity', 'abnormal diet', and 'not experienced medical check-up': 'high risk class', 'middle risk class', and 'low risk class'. Most of the participants belong to 'middle risk class'. Country of origin, age, length of stay, number of children, work status, health insurance status, and unmet health care needs were associated with problematic health behaviors in middle risk health behavior class. Conclusions: Health promotion and intervention programs for marriage immigrant women and their family members need to consider the health behavior patterns of physical inactivity, abnormal diet and no medical check-up and develop multiple behavior intervention with pre-existing program modification.