• Journal of Korean Society of Occupational and Environmental Hygiene
• /
• v.31 no.3
• /
• pp.237-254
• /
• 2021

Objectives: This study aims to classify the latent profiles in the exposure to hazard factors of salaried employees and test the determinants. Methods: Latent profile analysis(LPA) was conducted using data from the fifth Korean Working Conditions Survey(KWCS). 30,050 of salaried employees were the subjects of this study. After classifying the employees, multinomial logistic regression was used to test the determinants. Results: Salaried employees were classified with three latent profiles based on the exposure to the hazard factors. Employees included in class 1(32.8%) tend to experience low level of physical hazard factors, moderate level of psychological hazard factors, and high level of office work hazard factors. Employees included in class 2(61.8%) tend to be exposed to the moderate to high level of physical hazard factors, moderate to low level of psychological hazard factors, and low level of office work hazard factors. Employees included in class 3(5.4%) tend to experience high level of psychological and physical hazard factors, and moderate level of office work hazard factors. After classification, the demographic, health-, and employment-related variables were tested. Conclusions: This study clarified the features of each class, and proved that employees in class 3 are quite hazardous in that they are exposed to physical and psychological hazard factors much more frequently than other employees. Thus, this study can be used in predicting the high-risk employees and taking preemptive measures for preventing industrial accidents.

• Korean Journal of Social Welfare
• /
• v.66 no.3
• /
• pp.5-28
• /
• 2014

School absenteeism, one of the early warning signs of behavioral problems, has been known to be a complex and heterogeneous problem. However, much of the research assumes that school absentee adolescents comprise a homogeneous group. This study explored the heterogeneity of school absentee adolescents, based on school adaptation, to provide a more nuanced understanding of school absenteeism and examined predictive and risk factors associated with each typology of school absentee adolescents. Latent profile analysis was conducted using sample 477 middle school students who were reported absent in the previous year from the 3rd wave of Korean Children and Youth Panel Study. Multinomial logistic regression and ANOVA was also employed to examine the effects of predictive variables. As a result, three profiles were identified: low, middle, and high adaptive group. Group membership was found to be associated deferentially with gender, mental health, parenting neglect, delinquent friends, and delinquent behaviors. These findings propose more specific and targeted interventions designed to meet the needs and risk factors associated with the different typologies of school absentee adolescents.

• Journal of Korean Public Health Nursing
• /
• v.10 no.1
• /
• pp.12-22
• /
• 1996

Currently. exchanges of industrial workers between countries are more active than ever. and the problem of AIDS in connection with the operation of preventive educational programs has reached a point where the issue cannot be limited to native people alone any more. Based on such judgement. this research has been carried out to provide basic study materials by grasping the latent factors representing the difference between countries in the levels of right knowledge. attitude and behavior with respect to AIDS maintained by workers who have grown in different social and cultural living background. During the period from Apr. 1. 1995 to Jun. 30. questionnaires. written both in Korean and English, were distributed to Korean and Malaysian employees working at certain Korean video manufacturers. and the results of replies. given by 80 workers who were analyzed through matched sampling· method where ages and sex matched by country. were used as the research materials. The gathered materials were analyzed through the SPSS package t-test. ANOVA. factor analysis and multiple stepwise regression methods. and the following results were obtained. 1. The 2 extracted latent factors could be named 'common. social' knowledge factor and 'in-depth. psychological' knowledge factor respectively. 2. The percentile points of 'external. social' knowledge factor. in the case of Korean workers. howed 90.0 at average. a figure 13.75 points higher than those of Malaysian workers. 76.25. On the other hand. the percentile points of the 'in-depth. psychological'knowledge factor showed 70.80 at average in the case of Korean workers. a figure 7.47 points lower than those of Malaysian workers. 78.33. Meanwhile. the difference in percentile points between the 2 latent factors was 8.54 at average. indicating that the points of 'in-depth. psychological' knowledge factor was lower than those of the 'external. social' knowledge factor. 3. As for Korean workers. the percentile points of the 'in-depth. psychological' knowledge factor showed higher points in office workers than in non-office workers. and such variables exhibited in the position of workers can explain the $7\%$ of the latent factor. The percentile points of the 'in-depth. psychological' knowledge factors. in the case of Malaysian workers. showed higher points in groups who had religion than in groups who did not. and higher points in groups who obtained information from newspapers than in groups who obtained from televisions or other sources; and with these 2 variables. $26\%$ of this latent factor can be explained. The results. of analysis described so far suggest that while Korean workers possessed general level of knowledge on AIDS. they had low level of practical knowledge as far as its depth is concerned. and that they had social prejudice on patients as well as on the AIDS infection route. In addition. because the overall knowledge level of Malaysian workers. is lower than that of Korean workers. it suggests that separate programs intended for Malaysian workers are required prior to executing integrated programs.

• The Journal of Asian Finance, Economics and Business
• /
• v.7 no.6
• /
• pp.409-416
• /
• 2020

The research has been conducted to explore the combination of three intangible resources, including social capital, entrepreneurship, and resilience capability on the performance of State Capital Enterprises (SCEs) in Vietnam. Both qualitative and quantitative approaches are applied in the study. An in-depth interview of ten CEOs at SCEs in Vietnam was made to explore new indicators for the contextual latent variables in the research models. By employing the data from the authors' survey of 568 SCEs in Vietnam in 2019, using Cronbach's alpha, confirmatory factor analysis (CFA) and path analysis (SEM), the mechanism that social capital impacts on SCE performance has been analyzed. In addition to the direct role, social capital indirectly affects corporate performance through entrepreneurship and resilience capability. It was found that social capital has a larger impact on entrepreneurship than resilience capacity. However, the contribution of resilience capacity to the firm performance is much more than the entrepreneurship's in Vietnamese context. This study enriches the theory by proposing a measurement scale of the contextual latent variables as a result of in-depth interviews with experts using a qualitative analysis technique. In addition, the path analysis findings suggest practical implications for managers to effectively use their resources in SCEs.

• Korean Journal of Child Studies
• /
• v.37 no.6
• /
• pp.145-155
• /
• 2016

Objective: The purpose of this study was to investigate trajectories of child peer interaction and to compare the causal effects of their predictors, including child individual variables (i.e., gender, language ability, and cognitive ability) and teacher variables (i.e., teacher efficacy and teacher-child interaction). Methods: The participants of this study were 263 children and their teachers from the forth to sixth waves of longitudinal data from the Korean Children and Youth Panel Survey by the Korea Institute of Child Care and Education. The data was analyzed using Pearson's correlation and latent growth modeling. Results and Conclusion: The findings of this study are as follows: First, there was a linear decrease in child negative peer interaction over the course of 3 years, and significant individual differences were found in the trajectories (intercept and slope). Second, the predictors had significant casual effects on the trajectories of child negative peer interaction. The trajectories of child negative peer interaction involving girls, higher cognitive ability, and greater teacher-child interaction showed lower degree of intercept and a quicker decrease. Finally, the implications of findings are discussed.

• ETRI Journal
• /
• v.41 no.6
• /
• pp.771-781
• /
• 2019

We propose the VS³-NET model to solve the task of question answering questions with machine-reading comprehension that searches for an appropriate answer in a given context. VS³-NET is a model that trains latent variables for each question using variational inferences based on a model of a simple recurrent unit-based sentences and self-matching networks. The types of questions vary, and the answers depend on the type of question. To perform efficient inference and learning, we introduce neural question-type models to approximate the prior and posterior distributions of the latent variables, and we use these approximated distributions to optimize a reparameterized variational lower bound. The context given in machine-reading comprehension usually comprises several sentences, leading to performance degradation caused by context length. Therefore, we model a hierarchical structure using sentence encoding, in which as the context becomes longer, the performance degrades. Experimental results show that the proposed VS³-NET model has an exact-match score of 76.8% and an F1 score of 84.5% on the SQuAD test set.

• Human Ecology Research
• /
• v.59 no.2
• /
• pp.215-231
• /
• 2021

The study set out to identify the number of types of parental attitude profiles and their characteristics, as perceived by adolescents. The study also examined whether predictor variables (parent life satisfaction and self-esteem) influence parenting attitude profiles and whether these profiles, in turn, influence adolescent academic engagement and academic helplessness. The sample consisted of data on 2,590 adolescents and their parents from the 2018 Korean Children and Youths Panel Survey (KCYPS). The adolescents were in the first grade of middle school. Using Mplus, the author applied latent profile analysis to identify the parenting attitude profiles and predictor and outcome variables associated with these profiles. Three profiles were identified based on major features: (1) coercion-inconsistency (8.3%), (2) general (47.3%), and (3) warmth-autonomy support (44.4%). Parent life satisfaction and self-esteem predicted the classification of the profiles. Also, the parental profiles identified the level of adolescent academic engagement and academic helplessness. The findings have important implications for family policies and practices given the significance of parent psychological status, particularly life satisfaction and self-esteem, on their adolescent children's academic engagement and academic helplessness.

• Communications for Statistical Applications and Methods
• /
• v.30 no.6
• /
• pp.589-603
• /
• 2023

Deep generative models target to infer the underlying true data distribution, and it leads to a huge success in generating fake-but-realistic data. Regarding such a perspective, the data attributes can be a crucial factor in the data generation process since non-existent counterfactual samples can be generated by altering certain factors. For example, we can generate new portrait images by flipping the gender attribute or altering the hair color attributes. This paper proposes counterfactual disentangled variational autoencoder generative adversarial networks (CDVAE-GAN), specialized for data attribute level counterfactual data generation. The structure of the proposed CDVAE-GAN consists of variational autoencoders and generative adversarial networks. Specifically, we adopt a Gaussian variational autoencoder to extract low-dimensional disentangled data features and auxiliary Bernoulli latent variables to model the data attributes separately. Also, we utilize a generative adversarial network to generate data with high fidelity. By enjoying the benefits of the variational autoencoder with the additional Bernoulli latent variables and the generative adversarial network, the proposed CDVAE-GAN can control the data attributes, and it enables producing counterfactual data. Our experimental result on the CelebA dataset qualitatively shows that the generated samples from CDVAE-GAN are realistic. Also, the quantitative results support that the proposed model can produce data that can deceive other machine learning classifiers with the altered data attributes.

• Human Ecology Research
• /
• v.55 no.1
• /
• pp.13-26
• /
• 2017

This study investigated the latent groups depending on married working women's work-family spillover. The effects of factors that determine mental health subgroups and differences were also analyzed. Mixture modeling was applied to the Korean Longitudinal Survey of Women & Families to achieve the research objectives. The major findings of this study were as follows. First, there were four subgroups that could be defined according to the work-family spillover: mid-level spillover group (mid-positive and mid-negative spillover group), high-level spillover group (high-positive and high-negative spillover group), low-level spillover group (low-positive and low-negative spillover group), and high-negative and low-positive spillover group. Second, the results of mixture regression analysis to test the effect of eco-system variables showed that age, academic background, non-traditional family value, number of children, work hours, wage income, and availability of the maternity leave were significant determinants of the latent groups. The probability of classifying in the high-negative and low-positive spillover group increased when women showed a lower academic background and wage income, higher number of children and older age, and longer work hours than others. Third, the high-level spillover group, and the high-level spillover group showed the lowest stress and the lowest depression; however, the low-level spillover group reported the highest stress and the highest depression. Implications, limitations, and future directions were discussed based on the results.

• Journal of Hydrogen and New Energy
• /
• v.29 no.1
• /
• pp.19-24
• /
• 2018

Fuel cells are used to generate electricity with a reformer. In particular, methanol has various advantages among the fuels for reformer. Methanol steam reformer devices can efficiently supply hydrogen to PEM fuel cell. This study investigated the optimal operation conditions of a methanol steam reforming process. For this purpose, aspen HYSYS was used for the optimization of reforming process. The optimal operating condition could be designed by setting independent variables such as temperature, pressure and steam to carbon ratio (SCR). The optimal temperature and steam to carbon ratio were $250-270^{\circ}C$ and 1.3-1.5, respectively. It is advantageous to operate at a pressure of 15-20 barg, considering the performance of the hydrogen purifier. In addition, a heat exchange network was designed to supply heat constantly to reformer through the latent heat of steam.