首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到17条相似文献,搜索用时 144 毫秒
1.
本文目的是介绍一种可以考察各级交互作用的多因素设计类型,即析因设计。通过介绍两个实际例子,总结出析因设计的八个特点及优点、缺点。针对实际应用时可能面临的情况或问题,人们可能会对析因设计进行改造,由此产生了析因设计的三种变形,即"含区组因素的析因设计""分式析因设计"和"多因素非平衡组合试验"。最后一种情形,是实际工作者常会使用的,但在对定量评价指标进行差异性分析时需要对原先的分组进行合理拆分,再进行恰当的组合,构造出多种标准的设计类型,以利于分析者选择正确的统计分析方法。  相似文献   

2.
本文目的是介绍一种无法考察交互作用的多因素设计类型,即嵌套设计。通过介绍四个具体实例,总结出两种不同的嵌套设计:(1)试验因素之间具有自然属性上的嵌套关系;(2)试验因素对评价指标的影响存在主次关系。通过对实例的解读,揭示嵌套设计具有一个其他多因素设计所缺乏的明显特点,即因素的水平是"纵向连接"而不是"横向组合"的;并从正确合理应用的角度考量,指出了使用嵌套设计时应注意的事项。  相似文献   

3.
本文目的是介绍与均值比较有关的6个基本概念、计算方法、一个医学试验实例及其SAS实现。基本概念包括参数检验与非参数检验、前提条件、设计类型、交互作用以及方差分析的基本思想。计算方法涉及单因素两水平设计定量资料t检验、单因素多水平设计定量资料方差分析和含区组因素的析因设计定量资料方差分析。医学试验实例涉及如何考察可待因和针刺对男性受试者术后牙痛的影响。本文给出了采用SAS处理实例中定量资料的全过程,包括单因素分析、两因素分析和三因素分析,对输出结果作出解释,并明确给出合理选择统计分析方法进行定量资料差异性分析的关键点。  相似文献   

4.
本文目的是介绍拉丁方设计定量资料一元方差分析的计算公式和SAS实现。拉丁方设计可分为一般拉丁方设计与希腊拉丁方设计两大类,前者可以用于一个试验因素和两个区组因素的试验场合;后者可用于两个试验因素和两个区组因素的试验场合。事实上,还可以按是否进行重复试验和区组因素是否为单个体型,对拉丁方设计做进一步细分。一般来说,在对拉丁方设计定量资料进行方差分析时,除了应满足“独立性、正态性和方差齐性”的要求外,还要求所有因素之间的交互作用不存在或可以忽略不计。当定量资料不满足前述提及的前提条件时,建议采用混合效应模型建模并求解,或者基于广义估计方程方法求解方差分析模型中参数的估计值。  相似文献   

5.
本文目的是介绍一种广泛应用于生物医药学和临床试验研究中的多因素试验设计类型,即重复测量设计。根据试验分组因素数目和重复测量因素数目的不同,此设计类型具有多种不同的亚型。此设计类型特别适用于需要考察接受不同处理的受试对象随着时间的推移、定量指标取值的动态变化规律。此设计类型的一个突出特点是定量指标在不同时间点之间具有不相等的相关性,在对其定量资料进行差异性分析时,需要采用多种不同的方差协方差结构的方差分析模型,并依据特定的判定标准,从中选择最合适的方差分析模型给出最终的分析结果。  相似文献   

6.
本文目的是介绍一种特殊的处理多因素试验设计一元定量资料差异性分析方法,即结合分析法。通常情况下,处理多因素试验设计一元定量资料应采用方差分析。但是,此方法无法给出各对影响因素"重要性"的评价,也无法给出因素各水平的"效用值",更无法给出"理想试验点"。本文通过对一个实例的全面解析,显示了SAS中的TRANSREG过程具有很强且多样性的变量变换能力,它集方差分析、回归分析和结合分析于一体,能够很好地处理不符合传统统计学要求的复杂资料,能够实现前述期望达到的目的。  相似文献   

7.
本文目的是介绍平衡不完全区组设计定量资料一元方差分析的模型、计算公式和SAS实现。在计算中,涉及两个检验统计量,即FAFB。其中,A代表试验因素,B代表区组因素(即重要的非试验因素)。一般来说,假定平衡不完全区组设计中的两个因素之间不存在交互作用或交互作用可以忽略不计,故不需要评价交互作用项是否具有统计学意义,因此,在两因素各水平组合下可以不做重复试验。本文借助SAS软件,对两个实例进行平衡不完全区组设计定量资料一元方差分析,并给出计算结果,作出统计结论和专业结论。  相似文献   

8.
本文目的是介绍与单因素两水平和多水平设计定量资料多元方差分析有关的基本概念、计算方法、两个医学实例以及SAS实现。基本概念包括试验因素与属性因素、方差分析与协方差分析、自变量与协变量、矩阵与行列式;计算方法涉及一般统计量、4个多元方差分析检验统计量和多个均值向量的两两比较;两个医学实例分别为“2种脱机方式下患者脱机前有关定量指标的测定结果”和“3种不同类型地区的8岁男童生长发育调查数据”。借助SAS对两个实例中的定量资料分别进行多元方差分析和协方差分析,并对如何合理选取协变量进行阐释和讨论。  相似文献   

9.
本文旨在介绍两种多因素试验设计方法,即正交设计与均匀设计。由于析因设计中的试验点数目过多,在既要较少试验点数目,又要能比较准确地揭示多因素对评价指标影响规律的要求之下,宜选择正交设计或均匀设计。正交设计具有"试验点均匀分散、整齐可比且可以事先安排少数交互作用项"的特点,因此,结果的可信度较高,只需采用方差分析处理定量资料;均匀设计具有"试验点极其均匀分散、正交性在一定程度上受到破坏,不能事先安排交互作用项"的特点,因此,允许很少数目的试验点,但需采用多重回归分析处理资料且分析结果不唯一。  相似文献   

10.
如果试验涉及两个试验因素,当各因素在试验中同时实施且所处的地位基本平等,两个因素之间存在一级交互作用,且需要加以考察时,所采用的一种试验设计类型就是两因素析因设计[1].若观察指标只有一个且是定量的,相应的资料类型即为两因素设计一元定量资料,这时我们可以采用两因素设计一元定量资料方差分析处理该资料.本期讲座,我们向读者介绍两因素析因设计一元定量资料方差分析时样本含量与检验效能估计.  相似文献   

11.
本文目的是介绍一种无法考察交互作用的多因素设计类型,即交叉设计。通过详细介绍与交叉设计类型对应的"主要内容""操作方法"和"设计变形",全面展示了这种设计类型的核心内容、要领和注意事项,为用户正确、合理地选用交叉设计类型创造了有利的条件。  相似文献   

12.
Gloster  A. T.  Wittchen  H. U.  Einsle  F.  H&#;fler  M.  Lang  T.  Helbig-Lang  S.  Fydrich  T.  Fehm  L.  Hamm  A. O.  Richter  J.  Alpers  G. W.  Gerlach  A. L.  Str&#;hle  A.  Kircher  T.  Deckert  J.  Zwanzger  P.  Arolt  V. 《European archives of psychiatry and clinical neuroscience》2009,259(2):155-166

Cognitive behavioral therapy (CBT) is efficacious for panic disorder with agoraphobia (PD/A). Nevertheless, the active ingredients of treatment and the mechanisms through which CBT achieves its effects remain largely unknown. The mechanisms of action in CBT (MAC) study was established to investigate these questions in 369 patients diagnosed with PD/A. The MAC study utilized a multi-center, randomized controlled design, with two active treatment conditions in which the administration of exposure was varied, and a wait-list control group. The special feature of MAC is the way in which imbedded experimental, psychophysiological, and neurobiological paradigms were included to elucidate therapeutic and psychopathological processes. This paper describes the aims and goals of the MAC study and the methods utilized to achieve them. All aspects of the research design (e.g., assessments, treatment, experimental procedures) were implemented so as to facilitate the detection of active therapeutic components, and the mediators and moderators of therapeutic change. To this end, clinical, behavioral, physiological, experimental, and genetic data were collected and will be integrated.

  相似文献   

13.
The behavioral assessment of pain is essential for the analysis of pain mechanisms and the evaluation of analgesic drugs. The formalin test is one of such methods widely used as a model of injury-induced pain in rodents. This test is manually demanding and the recording of results is left to the subjectivity of the experimenters. Thus we developed a novel automated method to estimate the pharmacological response in formalin-induced licking behavior in rats using a multicolor detection technique. Two color markers were preliminarily applied to rats-yellow dye on the mouth and fluorescent green tape on the right hind paw. Behaviors of the animals were recorded from both above and below the subject, by a dual-view digital video camera system. After injection with formalin into the hind paw, rats exhibited a biphasic display of licking behavior. Licking time was measured by the sum of frames where the distance between these markers was less than an appropriate threshold of distance (TD). The split-plot analysis of variance demonstrated that the sum of squares of differences in licking time between manual and automated measurement was minimized when TD = 20mm. In addition, frames in which moving velocity of these markers is less than 2.5mm/s was neglected for calculation in order to eliminate sedative effect on the recorded data. On these conditions, subcutaneous administration of morphine in rats dose-dependently decreased formalin-elicited nociceptive responses. These results suggest that under optimal conditions the automated technique when applied to pharmacological studies are more reliable and efficient than if they are manually recorded.  相似文献   

14.
背景:在促进骨折修复研究中,应用外源性生长因子极不稳定,且造价高不适宜广泛推广,如何有效地促进内源性生长因子的表达将是值得深入研究的课题。 目的:观察生骨注射液对骨折愈合过程中内源性碱性成纤维细胞生长因子和血管内皮生长因子表达的影响。 设计、时间及地点:完全随机分组设计,对照实验,于2007-07/2008-08在华中科技大学同济医学院附属同济医院骨科实验室完成。 材料:清洁级3月龄雄性SD大鼠60只。生骨注射液由当归、土鳖虫、淫羊藿等药物组成,由湖北省中医院提供,质量浓度为1 g/L。 方法:建立SD大鼠胫骨干骨折愈合模型,分成实验组和对照组各30只,分别在骨折断端等量注射生骨注射液和生理盐水,每两日1次,0.2 mL/次。两组分别于术后第1,2,3,4,5,6周时各处死5只大鼠,取材。 主要观察指标:采用免疫组织化学方法,检测两组大鼠骨折愈合过程中不同阶段骨痂组织中碱性成纤维细胞生长因子、血管内皮生长因子蛋白的表达及差异情况。 结果:免疫组织化学染色显示,各时间点实验组骨痂组织中碱性成纤维细胞生长因子、血管内皮生长因子表达量及阳性定位均明显强于对照组。 结论:生骨注射液能够增加骨折愈合过程中碱性成纤维细胞生长因子、血管内皮生长因子在骨痂组织中的表达,这可能是生骨注射液促进骨折愈合的机制之一。  相似文献   

15.
The objective of this systematic review was to compare the results of studies of Internet-based eating disorder prevention programs. Eight electronic bibliographic databases, three key journals, and study reference lists were searched. This method yielded five published studies: four experimental and one quasi-experimental. A meta-analysis of the study results indicated no statistical significance for pooled study outcome data. No robust evidence exists on the impact of Internet-based prevention strategies on eating disordered symptomatology and on putative factors that contribute to eating disorder development. Further intervention development and consideration to study design are needed to move this field forward.  相似文献   

16.
The objective of this systematic review was to compare the results of studies of Internet-based eating disorder prevention programs. Eight electronic bibliographic databases, three key journals, and study reference lists were searched. This method yielded five published studies: four experimental and one quasi-experimental. A meta-analysis of the study results indicated no statistical significance for pooled study outcome data. No robust evidence exists on the impact of Internet-based prevention strategies on eating disordered symptomatology and on putative factors that contribute to eating disorder development. Further intervention development and consideration to study design are needed to move this field forward.  相似文献   

17.
Two profoundly retarded adolescents were punished for engaging in stereotyped self-stimulatory behavior. A within subjects reversal design was used. One adolescent also experienced a DRO condition to reduce his self-stimulatory behavior. Both participants exhibited what could be interpreted as negative side effects of the punishment condition. For one adolescent this was seen in experimenter directed aggression, while for the other the side effect was a decrease in social behavior. These behaviors were not seen in the other experimental conditions. The data suggest that punishment may result in negative side effects and when it is the treatment of choice, a careful, systematic monitoring of a variety of behaviors should be made. In this way any deleterious effects of the procedure are more likely to be detected.  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号