SPSS软件进行随机区组设计非参数检验的多重比较

[目的]随机区组设计多组样本资料非参数秩和检验后,在SPSS软件下实现各组样本的多重比较.[方法]采用秩转换技术结合随机区组设计的方差分析进行多重比较.[结果]对同一数据资料进行处理,所得结论相同.[结论]SPSS软件可实现非参数检验的多重比较.     

偏最小二乘回归的原理及应用

[目的]探讨偏最小二乘回归的理论及其应用。[方法]应用医学实例计算,对偏最小二乘回归与主成分回归及一般最小二乘回归进行比较。[结果]偏最小二乘回归对数据的拟合度优于主成分回归和一般最小二乘回归法。[结论]偏最小二乘回归适用于处理有多重共线性的资料,当解释变量个数多、样本量少时尤为有效,是稳健的数据“软”建模的统计方法。     

单因素重复测量设计与随机区组设计资料处理异同解析

目的探讨单因素重复测量设计与随机区组设计资料的方差分析的关系。方法利用文献数据^[1,2]，拟对资料进行单因素重复测量设计和随机区组设计2种处理，比较2种处理结果的异同。结果单因素重复测量设计资料在满足球对称的条件和正态对称条件下，与误用随机区组设计处理结果相同；尤其在球不对称且F值＞＞F界值时，采用随机区组设计处理与单因素重复测量分析校正前后的结果，也会出现一致情况。结论2种方法都是单因素试验的方差分析，但2种设计的方法和分析目的均不同，不可混同。     

临床试验中多终点期中分析方法研究

[目的]介绍临床试验中多终点期中分析方法。[方法]在多终点分析的基础上将单变量成组序贯期中分析方法进一步扩展,介绍多终点(变量)期中分析方法。[结果]多终点期中分析方法可以分为2大类,分别对每个终点进行单变量分析,同时对每个终点的检验界值进行校正;或利用所有观察终点构造总的检验统计量进行分析。[结论]确定单变量成组序贯期中分析界值的方法可以扩展用于多终点期中分析。     

单变量多因素观察性资料主要影响因素的逐步筛选

目的探讨如何用单变量分析的一般线性模型(GLM-Univariate)方法筛选“单变量多因素观察性资料”的主要影响因素。方法以住院患者病案首页数据中住院费用的影响因素分析为例,比较逐步回归方法与GLM-Univari-ate方法的优劣。结果逐步回归方法可能导致的主要问题是庞杂的哑变量处理和计算结果对哑变量赋值方法的高度依赖性,且不能分析因素的整体效应。GLM-Univariate不仅能克服上述问题,而且可以通过方差分析表中各因素(协变量)的P值和效应值η2大小,逐步筛选出主要影响因素和协变量,并建立相应的回归方程。结论GLM-Univariate过程可以完成单变量多因素观察性资料主要影响因素的筛选,并能够分析因素之间和因素与协变量间的交互效应,较逐步回归方法更为适宜。     

二阶段交叉设计资料的方差分析在SPSS中的实现

[目的]探讨二阶段交叉设计资料的方差分析在SPSS中的分析方法. [方法]通过构造统计模型,并结合实例给出SPSS程序来实现二阶段交叉设计资料的统计分析. [结果]给出了完全随机设计模型对处理因素和时间顺序因素进行比较的具体做法.[结论]二阶段交叉设计资料的方差分析在SPSS中可以很好的实现.     

成都市流动人口依赖性物质使用现状及其影响因素

[目的]探讨成都市流动人口依赖性物质的使用现状及其影响因素,为今后依赖性物质使用的预防提供依据. [方法]采用现况调查,应用方差分析、卡方检验和logistic回归分析的方法,对"成都市流动人口健康调查"资料进行单因素分析和多因素综合分析. [结果]成都市流动人口以酒精和烟草的使用所占比例最高,分别为26.10%和25.30%,单因素分析和多因素Logistic回归分析结果表明,性别、婚姻和每周工作时间为影响流动人口烟草、酒精和毒品使用的共同危险因素(P值均<0.05),焦虑心理症状是影响其毒品使用的危险因素(P<0.05).[结论]加强职业技能培训,提高劳动技能,缩短每周工作时间,做好依赖性物质使用前的预防工作,消除社会歧视等措施对流动人口依赖性物质使用的控制具有重要意义.     

近5年国内不同医学类核心期刊中统计方法应用的分析

[目的]了解各类医学期刊中统计方法的应用情况,掌握目前国内医学核心期刊中统计方法应用的动向,引导广大医务工作者正确使用常用的统计学方法。[方法]通过分层随机抽样的方法,对近5年内临床、药学、内科学、外科学等9类国内核心医学期刊进行调查,共调查文献2385篇。调查内容包括文献的资料类型、统计学方法及统计软件的应用等情况。[结果]在9类医学期刊的统计学方法应用中,相对数指标描述的占22.7%,均数±标准差描述的占27.9%,统计图\表描述的占31.5%;使用t检验占25.9%、方差分析占10%、卡方检验占21%、Fisher概率占1.1%、秩和检验占2.1%、Ridit分析占1.1%、两变量直线相关与回归占10.1%、多变量分析占1.0%。[结论]国内不同医学类核心期刊中,t检验最常用,其次为卡方检验和方差分析,且不同类型的医学期刊常用的统计方法不尽相同。     

方差分析和混合线性模型在重复测量数据中的应用探讨

[目的]通过混合效应线性模型与重复测量的方差分析在重复测量资料中的应用比较,说明两种方法右处理重复测量资料时的特点.[方法]对四川大学华西医院精神科就诊的139名患者进行随访调查研究,对调查表中的抑郁量袭分数分别运用重复测量的方差分析模型和混合线性模型处理.[结果]混合效应模型和重复测量的方差分析都是处理重复测量资科的重要的统计方法,前者在选择协方差结构下可对重复测量资料的固定效应和随机效应参数及协方差矩阵进行参数估计和统计检验,后者只能对重复测量资料的固定效应做出统计推断.[结论]混合效应模型在处理临床中的重复测量的资料比重复测量的方差分析,对资料的要求更宽松,所得的结果更加客观.     

实际应用中方差分析与秩和检验结果比较

在医学研究中，常常碰到单因素（完全随机设计）多个样本的比较。常用的比较分析方法是方差分析。但作这些分析时要求资料满足正态分布且各组方差齐性，否则需要进行变量转换或运用非参数统计方法。但在很多实际资料分析中，并没有对资料的正态性和方差齐性进行检验判断，就直接套用方差分析，这样分析得出的结果有时会与真实结果产生偏差，从而引起误导。本文在慢性病经济负担分析中运用了方差分析与秩和检验两种方法进行分析，并对两者分析结果进行比较，以期在实际运用中阐述两种统计方法的区别。     

A note on ANOVA assumptions and robust analysis for a cross-over study

Analysis of variance (ANOVA) methods are usually applied to analyse continuous data from cross-over studies. The analysis, however, may not have appropriate type I error when certain assumptions are violated. In this paper, we first clarify a conventionally minimum set of assumptions that validate the F-tests of ANOVA models for cross-over studies. We then provide a practical verification/remedy procedure based upon the theoretical developments. By applying the verification/remedy procedure, more robust analysis results can be expected from the ANOVA models.     

Examining costs of chronic conditions in a Medicaid population

PURPOSE: To compare health care costs and their components in patients with chronic illnesses. DESIGN: Quasi-experimental retrospective database analysis of an integrated state-Medicaid dataset. METHODS: Nine chronic illnesses and 28 two-disease combinations were evaluated in 284,060 patients. Dependent variables were total cost and the component costs (hospital, physician, home health and medical supplies, and pharmacy). Statistical analysis included analysis of variance (ANOVA) and multiple analysis of variance (MANOVA). RESULTS: The nine chronic illnesses studied were: psychosis, depression, cardiovascular illness, congestive heart failure, diabetes, acid peptic illness, respiratory illness/asthma, hypertension, and anxiety. Psychosis and depression patients had the highest mean yearly costs at $6,964 and $5,505, respectively. Highest component costs were mental health practitioners for psychosis and hospital costs for depression. All other conditions had significantly lower yearly costs. Component costs consisted primarily of pharmacy and hospital costs. Psychosis was a component in 5 of the 7 most costly chronic-disease concurrences. The highest disease-concurrence mean cost was for psychosis and depression ($18,318). CONCLUSIONS: The unique resource needs of different chronic illnesses should be considered in benchmarking and evaluating chronic-disease management programs.     

如何深入挖掘患者满意度数据

为了弥补国内患者满意度研究中普遍存在的对数据挖掘不够充分的缺陷,提出4种客观、简便、可操作性较强的结果分析方法:相关分析、方差分析、判别分析和象限图分析;并通过在浙江省l家三级综合性医院中的应用研究,证明这些方法的适用性和对医疗服务质量改进的指导作用.     

The scaling of restraint and the prediction of eating

The most frequently used measure of dietary restraint in controlled experiments involving arduous dieters is the Revised Restraint Scale (Herman & Mack, 1975). To date, researchers who have used the measure in controlled experiments have created a “high” versus “low” artificial dichotomy based on a median split of scores on the scale, apparently in an attempt to apply a standard analysis of variance model. The present study replicates a basic ice cream taste test experiment. It compares study results that emerge from using the Revised Restraint Scale as an artificial dichotomy (“factor”) versus a continuous independent variable in a general linear model. Results showed that an important joint effect emerged when restraint was treated as a continuous variable, which would likely have gone undetected using standard ANOVA, and that a greater proportion of variance could be accounted for in the continuous variable model. Researchers to date may have failed to optimally outline the nature of the relationship between dietary restraint and other variables that have been of interest.     

USAT: A Unified Score‐Based Association Test for Multiple Phenotype‐Genotype Analysis

Genome‐wide association studies (GWASs) for complex diseases often collect data on multiple correlated endo‐phenotypes. Multivariate analysis of these correlated phenotypes can improve the power to detect genetic variants. Multivariate analysis of variance (MANOVA) can perform such association analysis at a GWAS level, but the behavior of MANOVA under different trait models has not been carefully investigated. In this paper, we show that MANOVA is generally very powerful for detecting association but there are situations, such as when a genetic variant is associated with all the traits, where MANOVA may not have any detection power. In these situations, marginal model based methods, however, perform much better than multivariate methods. We investigate the behavior of MANOVA, both theoretically and using simulations, and derive the conditions where MANOVA loses power. Based on our findings, we propose a unified score‐based test statistic USAT that can perform better than MANOVA in such situations and nearly as well as MANOVA elsewhere. Our proposed test reports an approximate asymptotic P‐value for association and is computationally very efficient to implement at a GWAS level. We have studied through extensive simulations the performance of USAT, MANOVA, and other existing approaches and demonstrated the advantage of using the USAT approach to detect association between a genetic variant and multivariate phenotypes. We applied USAT to data from three correlated traits collected on 5, 816 Caucasian individuals from the Atherosclerosis Risk in Communities (ARIC, The ARIC Investigators [ 1989 ]) Study and detected some interesting associations.     

Relationship between Myers-Briggs psychological traits and use of course objectives in anatomy and physiology

The results from the Myers-Briggs Type Indicator (MBTI), which identified preferred psychological traits for 131 nursing students, were compared to their usage levels of course objectives in an undergraduate course in anatomy and physiology. The three usage levels (user, occasional user, and nonuser) were also compared to exam scores in the course, overall grade point averages (GPA) in first-year nursing, and the various psychological traits measured by the MBTI. A multivariate analysis of variance (MANOVA) indicated that users of objectives achieved significantly higher exam scores and maintained a higher GPA than occasional and nonusers. The MANOVA also indicated that users of course objectives preferred a sensing judging modality, whereas nonusers preferred an intuiting perceiving style to guide their studying and learning.     

A needs assessment investigation of continuing medical education in risk management

Training of physicians in risk management has become an important task for continuing medical education (CME) directors, particularly in states where risk management education is required for licensure. Physicians in Florida who completed an introductory program in risk management were surveyed to determine future topics for risk management education. Physicians identified desired topics using a five-point Likert-type scaling (5 = high priority; 1 = low priority) and rated 55 topics within 10 major subject areas. The means, standard deviations, and rank order of topics within the subject areas are reported. Multivariate analysis of variance (MANOVA) revealed no significant differences between surgeons and non-surgeons with respect to the subject areas. A significant difference, however, was shown, between academic physicians and non-academic (private practice) physicians; non-academic physicians gave significantly higher scores to six subject areas: malpractice, medical records, acts or omissions constituting negligence, negligence/malpractice actions, defenses, and risk management programs. The results suggest that it is not necessary to plan separate risk management education programs for physicians according to their specialty; one should note, however, the higher priorities given by non-academic physicians and consider these differences when planning advanced risk management programs.     

裂-裂区设计及其方差分析

目的介绍裂-裂区设计及其方差分析。方法结合实例介绍裂-裂区设计中实验单位、实验因素级别的正确识别和合理安排,以便正确选择方差分析模型。结果只有正确识别裂-裂区设计中实验单位、实验因素的级别,才能获得正确的方差分析模型以及相应的分析结果。结论裂区及裂-裂区设计是采用区组化控制技术进行的多因素研究设计,是完全随机区组设计、拉丁方设计等的综合运用。     

Analysis of data from group-randomized trials with repeat observations on the same groups

This study used Monte Carlo simulations to evaluate the performance of alternative models for the analysis of group-randomized trials having more than two time intervals for data collection. The major distinction among the models tested was the sampling variance of the intervention effect. In the mixed-model ANOVA, the sampling variance of the intervention effect is based on the variance among group×time-interval means. In the random coefficients model, the sampling variance of the intervention effect is based on the variance among the group-specific slopes. These models are equivalent when the design includes only two time intervals, but not when there are more than two time intervals. The results indicate that the mixed-model ANOVA yields unbiased estimates of sampling variation and nominal type I error rates when the group-specific time trends are homogenous. However, when the group-specific time trends are heterogeneous, the mixed-model ANOVA yields downwardly biased estimates of sampling variance and inflated type I error rates. In contrast, the random coefficients model yields unbiased estimates of sampling variance and the nominal type I error rate regardless of the pattern among the groups. We discuss implications for the analysis of group-randomized trials with more than two time intervals. © 1998 John Wiley & Sons, Ltd.     

不等距重复测量设计方差分析用SAS和SPSS实现的对比

目的探讨不等距重复测量设计方差分析的条件及结果解释。方法用SAS和SPSS软件实现不等距重复测量设计方差分析。结果给出了方差分析条件、SAS和SPSS程序及主要结果解释。结论只有满足方差分析条件、正确合理解释结果,才能保证不等距重复测量设计方差分析的有效性。