[Non-parametric Bootstrap estimation on the intraclass correlation coefficient generated from quantitative hierarchical data].

This paper aims to achieve Bootstraping in hierarchical data and to provide a method for the estimation on confidence interval(CI) of intraclass correlation coefficient(ICC).First, we utilize the mixed-effects model to estimate data from ICC of repeated measurement and from the two-stage sampling. Then, we use Bootstrap method to estimate CI from related ICCs. Finally, the influences of different Bootstraping strategies to ICC's CIs are compared. The repeated measurement instance show that the CI of cluster Bootsraping containing the true ICC value. However, when ignoring the hierarchy characteristics of data, the random Bootsraping method shows that it has the invalid CI. Result from the two-stage instance shows that bias observed between cluster Bootstraping's ICC means while the ICC of the original sample is the smallest, but with wide CI. It is necessary to consider the structure of data as important, when hierarchical data is being resampled. Bootstrapping seems to be better on the higher than that on lower levels.

[Using multivariate multilevel logistic model to study the influential factors of attitudes related to the Punishment on Smoking, among the residents in Guangzhou].

OBJECTIVE: To explore the attitudes towards the Smoking Ban policy and the bias in different levels of related society sanctions so as to provide evidence for implementing anti-smoking measures in the public places. METHODS: A multi-univariate-multilevel logistic model was developed to find the relation between penalty attitudes and the covariates' effects. RESULTS: 74.6% (3647/4892) and 61.2% (2994/4892) of the residents agreed to punish either the managers of public places or to the smokers. The two kinds of attitude were associated (χ2=1253.45, P<0.0001). Residents being female, aged between thirty and fifty-nine years old, having had higher than college education, being non-current smokers, with high scores on knowledge of tobacco control, tended to have the attitude of punishing the public place managers and smokers. However, those persons exposed to everyday secondhand smoke or being non-current smokers, tended to punish the smokers in the generalized Wald test. CONCLUSION: There was strong evidence noticed that most of the residents in Guangzhou had supported the 'tobacco control penalty'. Factors as age, sex, level of education received, knowledge on hazards of smoking, being current smokers and under secondhand smoke exposure were the influencing factors. Non-current smokers and those who exposed to secondhand smoke everyday, would prefer to take different penalties policies.

[Multivariate and multilevel model analysis on factors that influencing the literacy of health among high school students in Guangdong province].

OBJECTIVE: To analyze the awareness on health among high school students and its influencing factors in Guangdong. METHODS: Multi-stage sampling and questionnaire "2009 health awareness survey of the Chinese citizens" developed by our Department of Health, were used. Data were analyzed by multivariate multilevel model under MLwinN 2.19 software. RESULTS: The mean scores on knowledge and ideas, behaviors and related skills among 1606 high school students of Guangdong province, were 69.08 ± 14.81, 60.05 ± 16.85 and 74.99 ± 21.17 respectively. Three items on health showed that they all related to each other and relations between grades (0.972, 0.715 and 0.855) were greater than the individuals (0.565, 0.426 and 0.438). CONCLUSION: Factors as students from outside the Pearl River Delta region or from the rural areas, being male, at general secondary schools, at grade one, with poor academic performance and more pocket money etc., had lower levels on those related information of health.