• Journal of Environmental Impact Assessment
• /
• v.25 no.2
• /
• pp.124-140
• /
• 2016

Using the Korea Labor Institute Panel Study(KLIPS), this study investigates the impacts of water pollution on life satisfaction in Korea. Panel random-effects ordered probit model is used to consider the ordered property of life satisfaction data and heterogeneity of panel data. The proxy variables to reflect the degree of water pollution are biochemical oxygen demand(BOD) and total phosphorus(TP). In addition to the environmental variables above, other determinants used in various studies on life satisfaction such as economic, social, and demographic characteristics are included. Estimation results show that water pollution is negative and significant for life satisfaction. Other indicators such as income, age, house ownership, gender, education are positively related while urban residence and own business are shown to be negatively related.

• The Korean Journal of Applied Statistics
• /
• v.37 no.6
• /
• pp.783-799
• /
• 2024

Longitudinal data are measured repeatedly over time from the same subject. Therefore, the repeated outcomes have correlations, and it is necessary to estimate the covariate effect on the response variable while explaining the correlations. In longitudinal ordinal data analysis, the covariate effect is estimated using generalized linear mixed models using a logit link function or a probit link function. In this paper, we review the generalized linear mixed models and marginalized models with the two types of link functions for longitudinal ordinal data analysis. Specifically, a Bayesian cumulative probit linear mixed model with the probit link function is used to analyze Korean workplace panel survey (WPS) data, which is longitudinal ordinal data. In the model, the correlation matrix is high-dimensional and positive definite, and it is estimated using the hypersphere decomposition. In the WPS data, corporate training participation rate is considered as a response variable. Assuming different correlation structures, several models are compared. For the most suitable model, some explanatory variables, the annual effect, profit sharing schemes status, average annual training hours per person, and labor union status, have effects on corporate training participation rate.