ABSTRACT
To explore the appropriate method in estimating relative risk (RR)/prevalence ratio (PR) related to non-independent datasets.The simulation datasets generated by computer and case study were analyzed by two generalized estimating equation (GEE) models to investigate and compare the related applicability.Both convergence effects of log-binomial-GEE model and Robust Poisson-GEE model were almost 100％.The estimation results of the two GEE models were both closer to the true value.95％CI coverage of the two GEE models increased along with the reduction of class aggregation or the increase of the number of categories.Robust-Poisson-GEE model seemed to be more stable and steady than the log-binomial-GEE.The two GEE models could correctly evaluate the effects of exposure on the outcome in the case study.Rarely,there appeared problems on convergence of Robust Poisson or log-binomial-GEE model,and the accuracy was high.Both models could be used to estimate the RR/PR on non-independent epidemiological data.
ABSTRACT
To compare two commonly used statistical approaches:the multilevel model and the latent growth curve model in analyzing longitudinal data. A longitudinal data set, obtained from the quality of life in patients with colorectal cancer after operation, was used to illustrate the similarities and differences between the two methods. Results from the study indicated that the latent growth curve modeling was equivalent to multilevel modeling with regards to longitudinal data which could yield identical results for the estimates of parameters. Multilevel model approach seemed easier for model specification. However, latent growth curve model had the advantage of providing model evaluation and was more flexible in statistical modeling by allowing the incorporation of latent variables. Both multilevel and latent growth curve models were suitable for analyzing longitudinal data with advantages on their own, they could be chosen by researchers under different situation to be chosen accordingly by researchers under different situation.
Subject(s)
Humans , Colorectal Neoplasms , General Surgery , Longitudinal Studies , Models, Statistical , Multilevel Analysis , Postoperative Period , Quality of LifeABSTRACT
To compare two commonly used statistical approaches:the multilevel model and the latent growth curve model in analyzing longitudinal data. A longitudinal data set,obtained from the quality of life in patients with colorectal cancer after operation,was used to illustrate the similarities and differences between the two methods. Results from the study indicated that the latent growth curve modeling was equivalent to multilevel modeling with regards to longitudinal data which could yield identical results for the estimates of parameters. Multilevel model approach seemed easier for model specification. However,latent growth curve model had the advantage of providing model evaluation and was more flexible in statistical modeling by allowing the incorporation of latent variables. Both multilevel and latent growth curve models were suitable for analyzing longitudinal data with advantages on their own,they conld be chosen by researchers under different situation to be chosen accordingly by researchers under different situation.
ABSTRACT
To compare two commonly used statistical approaches:the multilevel model and the latent growth curve model in analyzing longitudinal data. A longitudinal data set,obtained from the quality of life in patients with colorectal cancer after operation,was used to illustrate the similarities and differences between the two methods. Results from the study indicated that the latent growth curve modeling was equivalent to multilevel modeling with regards to longitudinal data which could yield identical results for the estimates of parameters. Multilevel model approach seemed easier for model specification. However,latent growth curve model had the advantage of providing model evaluation and was more flexible in statistical modeling by allowing the incorporation of latent variables. Both multilevel and latent growth curve models were suitable for analyzing longitudinal data with advantages on their own,they conld be chosen by researchers under different situation to be chosen accordingly by researchers under different situation.
ABSTRACT
Objective To investigate the influence of class lecture on health care of sex and production to the sexual attitudes of college students. Methods 930 college students were assessed by The Sexual Attitudes Questionnaire(SAQ) before and after taking part in the lesson of health care of sex and production. Results The score of SAQ of all subjects was lower after class lecture (9.44 ± 2.67 ) than before ( 10.78 ± 2.65 ), there was significant difference(P＜ 0.05 ). Compared to before class lecture , the scores of male and female college students were decreased after education(P＜0. 05 ). There was apparent difference between clinical and nonclinical students before and after class lecture ( clinical (99.70 ± 2.50) vs ( 8.30 ± 1.70), P ＜ 0.05; nonclinical (11.71 ± 2.67)vs (10.81 ± 2.23), P＜0.05). Conclusion Sexual knowledge was increased, and the sexual attitudes was tended to maturity of the students. It' s nessary to promote opening lecture of Health Care of Sex and Production in college students.
ABSTRACT
This article discusses how to improve teaching quality of health statistics from four main aspects: paying attention to training statistical thinking,paying attention to understanding basic concepts and methods,paying attention to modernistic teaching instrumentality,and paying attention to intercommunion between teachers and students.Practice indicates it facilitates impoving teaching quality.