• The Korean Journal of Applied Statistics
• /
• v.28 no.2
• /
• pp.269-279
• /
• 2015

This paper considers inferences of random effects. We show that the proposed confidence distribution (CD) performs well in logistic regression for random intercepts with small samples. Real data analyses are also done to identify the subject effects clearly.

• Korean Journal of Clinical Laboratory Science
• /
• v.56 no.3
• /
• pp.248-256
• /
• 2024

The purpose of this study was to identify the household air pollutants most significantly impacting hearing loss, using data from the 8th Korea National Health and Nutrition Examination Survey. Of the 1,980 participants, those with missing data were excluded from the study population, resulting in the final inclusion of 298 participants. Mild hearing loss is defined as a unilateral or bilateral pure-tone threshold average (PTA) of 26~40 dB, and moderate hearing loss is defined as a unilateral or bilateral PTA of 41 dB or higher. The mean of the PTAs for both ears is defined as the sum of the PTAs of the left and the right ears divided by 2. Complex samples multiple logistic regression analysis revealed that formaldehyde exposure independently significantly increased the risk of moderate hearing loss (odds ratio=1.050, P<0.001). Additionally, in the complex samples general linear model, toluene exposure (B=0.026, P<0.001) was found to be independently significantly associated with an increase in the mean of PTAs for both ears, where a higher mean of the PTAs for both ears indicates a decrease in hearing ability. These results suggest a strong association between hearing loss and the presence of toluene or formaldehyde in indoor air.

• Korean Journal of Social Welfare
• /
• v.62 no.4
• /
• pp.275-296
• /
• 2010

This study analyzed the state of battered women who drink at a dangerously high level and the factors that influence their hazardous drinking in order to promote their safety and their healthy adjustments. The researchers utilized the data from "Analysis of the Effectiveness of Intimate Partner Violence Offenders Correction and Treatment Program" funded by Ministry of Gender Equality, and thereby, sampled 124 battered women for interview over phone. The results of applying the WHO's standard for hazardous drinking based on frequency and quantity calculations showed that 15.7% of survey participants were classified as hazardous drinkers, either currently experiencing or having the risk of experiencing problems related to drinking. The result of carrying out a logistic regression analysis to identify the factors that influence hazardous drinking showed that household income and self-blame were factors that exerted statistically significant influence. Based on these results, clinical and practical intervention plans for helping battered women who drink hazardously were discussed.

• Journal of Digital Convergence
• /
• v.12 no.4
• /
• pp.367-378
• /
• 2014

The purpose of this study was to explore the factors related to internet addiction in college student of Korea. Methods: From November 19, 2012 until December 14, 2012, 300 college students from one of the universities located at K city were surveyed. This study is a cross-sectional survey of college students in K city, and data was collected through self-report questionnaires. Data was analyzed by Fisher's exact test, independent t-test to identify different according to addiction group, and logistic regression analysis in order to factors related to on health behavior, sexual behavior and mental health of among college students. According to results, factors related to risk of addiction were lower frequency of exercise (p<0.05), no eat breakfast for 2 day (p<0.05), porno magazine/porno video contact pornography (p<0.05), subjective happiness index (p<0.05), and suicidal Ideation (p<0.05). This study suggest that exercise, regular diet, important of reproductive health, subjective happiness index and Suicidal Ideation are associated with internet addiction in College student. It is necessary to develop some preventive programs for those of internet addiction.

• Journal of Preventive Medicine and Public Health
• /
• v.36 no.2
• /
• pp.137-146
• /
• 2003

Objectives : Impaired hearing is a prevalent occupational hazard, not only in industry, but also in the armed forces. In military life, noise has unusual characteristics, and constitutes a serious hazard to hearing. The aim of this study was to analyze the hearing threshold data in order to compare the hearing loss among shipyard workers, representing different workers, and a military service background. Methods : A cross-sectional audiological survey, combined with a questionnaire study, was conducted on a stratified random sample of 440 shipyard workers, with long-term exposure to noise. The employees were divided into four groups, according to their working and military service backgrounds, in relation to their exposure to noise. Results : As expected, the working and military noise exposure group (Group I) had significantly poorer hearing than the other groups. The high frequencies (2-8 kHz) showed the greatest difference in terms of poorer fearing in both ears. The prevalence of noise-induced hearing loss (NIHL) was highest in Group I. A logistic regression analysis was applied to determine the dependence of the NIHL in relation to age, smoking, drinking, working duration, ear protection, past history of ear diseases, and working and military sonics backgrounds, on the noise exposure The important factors found to be related to the NIHL, in relation to noise exposure were: age, work duration, and working and military service backgrounds. The adjusted odds ratio estimates for NIML in the right ear were 4.5 times greater (95% CI 1.7-11.6) for the military noise exposed group, and 7.9 times greater (95% CI 2.0-31.3) for the working noise exposed group than in the controls. The hearing thresholds at the pure-tone average and 4 kHz were significantly increased with age and work duration with both the working and military service backgrounds. Conclusions : From these results, specific preventive programs were planned, which should be assessed by epidemiological surveillance of the military noise exposed population.