结构方程模型在评价指标体系结构效度中的应用

目的 探讨结构方程模型在评价指标体系结构效度中的应用,并以2007年度首发基金立项函评指标体系为例,评价其结构效度.方法 随机将部分评审结果共1200项分为PART1与PART2两部分.PART1用于探索性因子分析以探索指标体系的结构,形成模型F1,并利用结构方程模型中的测量模型验证F1的结构效度.然后将修正系数与实际联合分析,将模型F1修正为模型F3.最后对PART2用验证性因子分析评价模型F3的结构效度.结果 最终模型F3的拟合指数结果显示模型的拟合优度较好.所有的路径系数、载荷系数及其他参数估计均有统计学意义.结论 结构方程模型是一种充分考虑专业理论、可靠的分析方法.它通过分析比较,选出相对满意的因子结 构方案,从而为研究提供合理的理论依据和专业结论.在评价指标体系结构效度的研究中引入结构方程模型具有很重要的现实意义.     

临床医学科技成果评价指标体系的实证研究

目的 以某省科技成果评审为例,对前期所建立的临床科技成果评价指标体系进行验证.方法 制定各评价指标的详细评分准则,由5位专家对某省2010年度20份科技成果申报材料进行模拟评分.采用加权求和法计算各评审项目的 总得分,考核指标体系的信度和效度.结果 指标体系的克朗巴赫α系数为0.91.各实测指标评分与总得分以及所属一级...     

呼吸道传染病健康素养综合评价体系的建立与评估

目的 筛选评价呼吸道传染病健康素养的指标,确定评价指标和维度的权重系数,建立并评估呼吸道传染病健康素养的综合评价体系.方法 依据性别、年龄、教育程度较均衡的配额抽样方法在北京市海淀区随机抽取了515名成人进行计算机辅助技术的问卷调查.通过难度、区分度、信度和效度检验筛选出测评指标;利用归一化法和变异系数法分别得到主观权重和客观权重,并利用综合指数模型最终构建评价体系.结果 共筛选39个指标,从知识、行为/行为意向、能力3个维度评价呼吸道传染病健康素养.各指标难度P均在0.5～0.9范围内,区分度D＞0.3,3个维度的内部一致性系数分别为0.782、0.657和0.842,赋权重后综合评价一致性信度达到0.872,二次探索性因子分析结果显示评价体系符合设定维度的划分(KMO=0.821,可解释方差为54.08％),以验证性因子分析进行验证,拟合优度良好( IFI=0.905,TLI=0.873,CFI=0.905,RMSEA=0.065).结论 本研究所开发的呼吸道传染病健康素养评价工具及体系的难易度、区分度、信度、效度均较好,可以用来测评公众的呼吸道传染病健康素养水平.     

基于验证性因子分析的中医工作满意度量表研制

目的研制中医医生工作满意度调查表,并对其进行评价。方法采用文献研究、专家咨询等方法研制中医医生工作满意度量表,采用Cronbach’sα系数、探索性因子分析和验证性因子分析评价调查表的信度和效度。结果中医医生工作满意度调查表共25个条目。调查表总体信度为0.952,说明问卷满意度量表内部一致性和可靠性好。各维度因子载荷均大于0.5,高于标准值0.4。经过因子载荷矩阵旋转提取出6个主成分,公因子累计贡献率为72.463%。运用验证性因子分析进行验证,修正后的模型各项指标均达到标准,显示探索性因子分析的结论基本可以接受,说明问卷具有良好的结构效度。同时各条目在其公因子的标准化载荷值均大于0.5,说明问卷具有良好的内敛效度。结论中医院中医医生工作满意度量表具有良好的信度和结构效度,可作为中医医生工作满意度的测评工具。     

流动人口SF-12生命质量量表信度、效度评价

目的 评估SF-12生命质量量表用于评价流动人口生命质量时的信度和效度.方法 采用内部一致性信度评估SF-12生命质量量表的信度,采用集合效度、区分效度和结构效度评估SF-12量表的效度.结果 SF-12生命质量量表用于评价流动人口生命质量时的内部一致性信度cronbach's α=0.84,各维度与总分的相关系数除了躯体活动功能(PF)=0.43,其余均>0.50;各维度的cranach's α系数均>0.70,且在删除相应维度后的cronbach's α系数均>0.70;8个维度的集合效度定标实验成功率均为100%,区分效度定标实验成功率均为100%;对量表的理论结构模型进行验证性因子分析,所得模型结构与原始假定一致,拟合指标结果为不规范拟合指数(NNFI)=0.95、比较拟合指数(CFI)=0.96、调整后的拟合优度指数(AGFI)=0.96、近似误差均方根(RMSEA)=0.06.结论 SF-12生命质量量表用于评价流动人口生命质量时具有较好的信度和效度.     

验证性因子分析在量表结构效度考核中作用

目的探讨验证性因子分析在新编量表结构效度评价中的作用。方法以乙型肝炎患者生存质量测定量表（QOL-HBV）为工具,用探索性因子分析（主成分分析、方差极大旋转法）和验证性因子分析2种方法评价量表的结构效度。结果探索性因子分析将量表归为8个因子,每个因子能得到合理解释[抽样适度测定值（KMO）0.883,解释的总方差为69.15%],将8个因子再进行因子分析,得到2个因子;用AMOS软件对8个因子进行验证性因子分析,得到二阶七因子模型,该模型的χ^2/DF为1.310,拟合优度指数（GFI）,调整拟合优度指数（AGFI）均〉0.90。结论探索性因子分析得出的因子间尚存在某种潜在的联系,且不能用探索性因子分析解释,用验证性因子分析可以弥补这缺陷,因此最好采用探索性因子分析与验证性因子分析相结合的方法来评价量表的结构效度。     

医教协同背景下临床医学专业学位研究生培养质量评价指标体系的评价

目的 构建用于评价临床医学专业学位研究生培养质量的指标体系,为临床专硕人才培养及考核评价提供测量工具。方法 对临床专硕研究生展开问卷调查,通过探索性因子分析评价指标体系结构效度。结果 两轮专家函询问卷回收率均为100%,专家积极性高,专家的平均权威程度系数为83.00%,经两轮Delphi法,指标条目由原来的76个变为58个,Kendall W由0.23提高为0.46（P<0.01）。指标体系总体的Cronbach α为0.96,其中各条目的Cronbach α为0.84～0.98,各维度的相关系数为0.32～0.66。探索性因子分析,提取6个公因子,累积方差贡献率为54.98%。结构方程结果显示,拟合效果较为理想（CFI、NFI、IFI均＞0.90,RMSEA均≤0.18）。高分组自评满意度得分均高于低分组,差异有统计学意义（P<0.001）,说明指标体系有良好的区分度。结论 构建的临床医学专业学位硕士研究生培养质量评价体系经检验信效度良好,可以为学校提高临床医学专业学位硕士研究生培养质量提供参考依据。     

北京市居民传染病健康素养指标体系研究

目的 建立和评价北京市居民传染病健康素养指标体系,为制定传染病健康素养综合评价指数奠定基础.方法 采用多阶段分层抽样方法抽取北京市6个区县18岁以上居民,通过问卷方法调查居民对传染病健康素养的3个维度(知识、行为、技能)的相关指标,根据答案给各指标赋值.应用专家咨询、难易度与区别度分析和验证性因子分析筛选出指标进入传染病健康素养指标体系,并通过计算分半信度、克朗巴赫α系数、验证性因子分析、Q型聚类分析等方法对指标体系的信度和效度进行评价.结果 共调查13 287人,回收有效问卷13 001份,有效应答率97.85％.经过专家咨询、难易度与区别度分析和验证性因子分析等方法筛选出30个指标进入传染病健康素养指标体系,包括知识性素养指标12个、行为性素养指标13个、技能性素养指标5个.传染病健康素养指标体系的分半信度为0.609,克朗巴赫α系数为0.777;验证性因子分析结果显示,所有标准化回归系数均有统计学意义(P＜0.01);Q型聚类将样本分为3类(分别有4819、3295和2405人),通过方差分析,发现各指标组间差异均有统计学意义(P＜0.001).北京市居民具备传染病健康素养的比例为9.9％.结论 研究构建的传染病健康素养指标体系具有较好的信度和效度,是评估北京市居民传染病健康素养的科学有效工具.     

均值结构模型在抑郁症病例-对照临床研究中的应用

目的 探讨如何利用均值结构模型实现潜变量数据的分组比较.方法 结合抑郁症病例-对照临床研究量表数据,利用LISREL软件实现参数估计,并以Δχ2(Δd)、RMSEA、NNFI、CFI等拟合优度指标评价拟合效果.结果 社会支持量表的验证性因子分析模型拟合优度不理想,但其均值结构模型拟合效果有所改善;特质应对方式问卷模型拟合优度检验比较理想.结论 采用均值结构模型进行抑郁症病例-对照分组比较,不仅获得量表结构效度及关联性结果,而且得到病例组与对照组潜变量均值的差异,结果符合专业解释.     

北京市居民传染病健康素养指标体系研究

目的 建立和评价北京市居民传染病健康素养指标体系,为制定传染病健康素养综合评价指数奠定基础.方法 采用多阶段分层抽样方法抽取北京市6个区县18岁以上居民,通过问卷方法调查居民对传染病健康素养的3个维度(知识、行为、技能)的相关指标,根据答案给各指标赋值.应用专家咨询、难易度与区别度分析和验证性因子分析筛选出指标进入传染病健康素养指标体系,并通过计算分半信度、克朗巴赫α系数、验证性因子分析、Q型聚类分析等方法对指标体系的信度和效度进行评价.结果 共调查13 287人,回收有效问卷13 001份,有效应答率97.85％.经过专家咨询、难易度与区别度分析和验证性因子分析等方法筛选出30个指标进入传染病健康素养指标体系,包括知识性素养指标12个、行为性素养指标13个、技能性素养指标5个.传染病健康素养指标体系的分半信度为0.609,克朗巴赫α系数为0.777;验证性因子分析结果显示,所有标准化回归系数均有统计学意义(P＜0.01);Q型聚类将样本分为3类(分别有4819、3295和2405人),通过方差分析,发现各指标组间差异均有统计学意义(P＜0.001).北京市居民具备传染病健康素养的比例为9.9％.结论 研究构建的传染病健康素养指标体系具有较好的信度和效度,是评估北京市居民传染病健康素养的科学有效工具.     

青春期学生逆反心理问卷编制与一般特征

目的探讨青春期学生逆反心理的结构并在此基础上编制量表。方法采用访谈法和半开放式问卷,结合相关理论获得青春期逆反心理维度,形成初始问卷,并分别对576名和600名青春期学生进行施测,通过探索性因素分析和验证性因素分析检验问卷信度效度等各项指标。结果青春期逆反心理主要由情绪逆反、认知逆反、行为逆反3个维度构成;验证性因素分析表明χ2/df=2.747,GFI=0.915,AGFI=0.872,NFI=0.836,NNFI=0.906,RMSEA=0.058,CFI=0.913,IFI=0.914;量表内部一致性系数为0.823,各维度内部一致性系数为0.761～0.836。结论该量表结构明晰,信效度均达到测量学指标。性别、年龄和家庭类型对青春期学生逆反心理存在影响。     

Zufriedenheit in der Arztpraxis aus Patientenperspektive

The purpose of the study was to develop and to validate a standardized patient centered quantitative instrument to assess patient satisfaction in ambulatory care, to be used i.e. in quality management and benchmarking activities. Item selection for the instrument was based on an extensive qualitative pilot study. A pre-final version (containing 35 items) has been tested with 2.800 patients from 56 ambulatory care practices (bases for analysis: n= 969). After psychometric evaluation a modified version has been tested with 3.000 patients from 60 ambulatory care physicians. Answers from 1.387 patients (response rate: 46 %, 62 % female, mean age 52 yrs.) were available for analysis. The questionnaire was validated by testing for factorial validity (confirmatory factor analysis), convergent and divergent validity as well as reliability (internal consistency, cronbachäs alpha). The final version of the questionnaire contains four dimensions: patient-physician-interaction (8 items), information (8 items), office organization(4 items), professional competence (3 Items). Confirmatory factor analysis confirms the four hypothesized dimensions perfectly (GFI: 0.997). Reliability ranged from.75 and.95. The correlation’s between the sub-scales ranges from.76 to 47, indicating that the dimensions are not fully independent. All sub-scales correlate positively with global assessments of satisfaction with the physician. With satisfactory to good results for validity and reliability testing, the final questionnaire might be used fruitfully in quality management in primary care.     

Concurrent and construct validity of the audit in an urban brazilian sample

AIMS: To assess the concurrent and the construct validity of the Alcohol Use Disorders Identification Test (AUDIT) in an urban Brazilian sample. METHODS: A random sample of 166 clients of a health management organization, participated in this study. They were visited in their households and completed a self-report questionnaire, which included the AUDIT. Later, they answered the alcohol-related disorders (ARDs) Section of the Composite International Diagnostic Interview. The receiver operating curve (ROC) was used to find the best cut-off point for ICD-10 diagnosis of ARDs. Confirmatory factor analysis was run to assess the construct validity. RESULTS: The ROC analysis showed the same cut-off point (7/8) for ICD-10 diagnosis of ARDs found in previous studies carried out in primary care settings, including in Brazil, with a sensitivity of 100% and a specificity of 76%. The confirmatory factor analysis suggested a two-factor structure. The first factor measured consumption and the second factor alcohol-related problems. CONCLUSIONS: The results supported the use of the self-reported version of the AUDIT in epidemiologic studies, and showed a similar cut-off point for detection of ARDs and hazardous drinking.     

WHOQOL-BREF用于评价HIV感染者生活质量的信、效度研究

  目的  评价世界卫生组织生活质量测定简表(World Health Organization quality of life scale-brief form questionnaire, WHOQOL-BREF)在测定HIV感染者生活质量中应用的信度和效度,比较不同特征HIV感染者的生活质量,为下一步评价感染者的生活质量和制定有效的关怀措施提供参考依据。  方法  采用方便抽样的方法在云南省红河哈尼族彝族自治州选择符合条件的HIV感染者。采用Cronbach’s α系数和分半信度评价其信度;探索性因子分析(exploratory factor analysis, EFA)和验证性因子分析(confirmatory factor analysis, CFA)检验其效度;单因素方差分析比较不同特征感染者在WHOQOL-BREF各维度上的得分情况。  结果  共获得410份有效问卷。WHOQOL-BREF及生理领域、心理领域、社会领域和环境领域的Cronbach’s α系数分别为0.901、0.779、0.819、0.729和0.743,分半信度分别为0.798、0.716、0.822、0.736和0.697。EFA共获取4个公因子,对总体方差的贡献率为54.542%;CFA结果显示最终拟合模型的主要指标χ2/df和估计误差均方根(root-mean-square error of approximation, RMSEA)分别为2.169和0.076;拟合优度模型指数(goodness of fit index, GFI)、比较拟合指数(comparative fit index, CFI)和Tucker-Lewis指数(Tucker-Lewis index, TLI)分别为0.820、0.861和0.844,均接近0.900,表示拟合结果可以接受。单因素方差分析结果显示,是否接受抗逆转录病毒治疗、民族、职业、文化程度和最近一次CD4+T淋巴细胞计数等不同特征感染者的生活质量有所不同。  结论  WHOQOL-BREF在测定HIV感染者的生活质量具有较好的信度和效度,不同特征感染者的生活质量有所差异,在制定有针对性的干预策略和措施时需加以考虑。     

Psychometric properties of the Farsi version of effort-reward imbalance questionnaire: a longitudinal study in employees of a synthetic fibre factory in Iran

Purpose

To assess the reliability and validity of the Farsi version of the effort-reward imbalance questionnaire (F-ERIQ) and to examine the responsiveness of the tool to changes over time.

Methods

A longitudinal study was carried out among 227 male employees of Iran Polyacryl Corporation. The F-ERIQ was developed through a forward–backward translation process that includes three scales of effort, reward and over-commitment (OC). Reliability and internal consistency of the F-ERIQ were assessed by split-half and Cronbach’s alpha coefficients. Confirmatory factor analysis, convergent and discriminant validity were conducted to evaluate construct validity. Depressive mood was used as an indicator for exploring criterion validity. The variations in mean scores over time for scales were regarded as measures of the responsiveness to changes.

Results

Baseline split-half correlations for effort, reward and OC were 0.53, 0.85 and 0.65, respectively; Cronbach’s alpha coefficients improved from 0.61 to 0.70 for effort, 0.85 to 0.88 for reward and 0.67 to 0.72 for OC. All of item-total correlations were higher than 0.23 and item-scales correlations were higher than 0.4. Although Values of Goodness of Fit Index and Adjusted GFI were higher than 0.9 and Root Mean Square Error of Approximation, Root Mean Square Residual and Standardized RMR were lower than 0.05, confirmatory factor analysis only confirmed the construct of the effort and OC. People with higher job stress were at higher risk of depressive mood (at least 3 times more). Overall, the mean score of effort, OC and ERI increase, and the figures decrease for reward among people who experience changes.

Conclusions

These findings provide evidence that the F-ERIQ is a reliable and valid instrument for assessing psychosocial stress at work among Farsi-speaking male employees. We propose that F-ERIQ be further evaluated across a variety of jobs and industries.     

Factor structure of the SOC scale 13-item version in Japanese university students

The purpose of the present study was to examine the construct validity of Antonovsky's (1987) sense of coherence scale (SOC) 13-item version, with a sample of Japanese university students. Subjects were 1,110 university students who completed a self-administered questionnaire including items on individual characteristics (age, sex), the Zung self-rating depression scale (SDS) and the SOC scale. Exploratory factor analysis of the SOC scale yielded an interpretable two-factor solution with "comprehensibility-manageability" and "meaningfulness". Confirmatory factor analysis demonstrated an a priori defined SOC second-order factor model composed of comprehensibility-manageability and meaningfulness factor to have an acceptable fit (chi-square=327.065, df=64, GFI=0.957, CFI=0.872, RMSEA= 0.061). Moreover, structural equation modeling showed that there are theoretically consistent relationships among individual SOC factors and depression. In particular, the meaningfulness factor proved better predictor of depression than the comprehensibility-manageability factor. These findings support the construct validity of the SOC scale and suggest that two SOC sub scores may provide clinically applicable information, rather than the total score.     

脑卒中患者健康行为评估工具的构建及信效度检验

目的 编制脑卒中患者健康行为问卷,并测定信效度。方法 在参考文献基础上编制脑卒中健康行为问卷,经过专家咨询确定问卷雏形,采用方便抽样抽取三所三级甲等医院381例住院脑卒中患者为研究对象,进行问卷调查,采用探索性及验证性因子分析对问卷信效度进行评价。结果 问卷总Cronbach’sα系数为0.807,总内容效度指数(S-CVI)为0.97,各条目内容效度指数(I-CVI)在0.8～1之间,经两次探索性因子分析,5因素包含的题项与编制问卷使用的概念框架大致符合,5因素累积方差贡献率为50.833%;模型拟合指数调整卡方(χ2/df)为1.168,近似误差均方根(RMSEA)为0.021,拟合优度指数(GFI)为0.942,比较拟合指数(CFI)为0.992,增量拟合指数(IFI)为0.992,各题项在各自潜变量上的负荷值都达到了显著水平,模型的各拟合指数达到了可接受的统计学标准。结论 该问卷具有较好的效度和信度,可为评估脑卒中患者健康行为提供依据。     

Validation of a questionnaire measuring patient satisfaction with general practitioner services

BACKGROUND: In order that patient satisfaction may be assessed in a meaningful way, measures that are valid and reliable are required. This study was undertaken to assess the construct validity and internal reliability of the previously developed Patient Satisfaction Questionnaire (PSQ). METHOD: A total of 1390 patients from five practices in the North of England, the Midlands, and Scotland completed the questionnaire. Responses were checked for construct validity (including confirmatory factor analysis to check the factor structure of the scale) and internal reliability. RESULTS: Confirmatory factor analysis showed that items loaded on the appropriate factors in a five factor model (doctors, nurses, access, appointments, and facilities). Scores on the specific subscales showed highly significant positive correlations with general satisfaction subscale scores suggesting construct validity. Also, the prediction (derived from past research) that older people would be more satisfied with the service was borne out by the results (F (4, 1312) = 57.10; p < 0.0001), providing further construct validation. The five specific subscales (doctors, nurses, access, appointments, and facilities), the general satisfaction subscale, and the questionnaire as a whole were found to have high internal reliability (Cronbach's alpha = 0.74-0.95). CONCLUSION: The results suggest that the PSQ is a valid and internally reliable tool for assessing patient satisfaction with general practitioner services.     

关系质量问卷中文自评版用于儿童师生关系研究的信效度

【目的】 考察关系质量问卷中文自评版用于儿童师生关系质量评估的信度和效度。 【方法】 采用分层整群抽样方法,从某小学及某中学中各抽取四个班共979名学生,回收有效问卷822份。随机抽取57人用于重测。 【结果】 关系质量问卷用于评估师生关系时的Cronbach α系数为0.84,重测信度为0.77。限定因子个数为2进行主成分分析解释变异量的39.86%。验证性因素分析的拟合指标χ²/df、RMSEA、GFI、AGFI、NNFI、CFI分别为3.42、0.09、0.78、0.74、0.89、0.90。 【结论】 关系质量问卷的信度、效度均良好,符号心理测量学要求,可用于今后国内的有关研究。     

大学生心理咨询倾向性问卷的编制

目的编制大学生心理咨询倾向性问卷。方法在查阅文献和前期访谈基础上编制问卷。进行3次大一学生的整群抽样,第一次调查315名大一学生用于初步项目分析和因素探索;第二次调查584名大一学生用于探索性因素分析和问卷信、效度检测;第三次调查101名大一学生用于验证性因素分析,以专业心理求助态度量表为效标,并进行重测。结果大学生心理咨询倾向性问卷共19个题目,包括心理咨询偏见、心理咨询信任、非专业心理求助途径、心理咨询污名4个因子。验证性因素分析表明所得模型拟合度可接受(χ2/df =2.677,GFI=0.932,AGFI=0.912,CFI=0.916,NFI=0.874,RMSEA=0.054)。总问卷内部一致性系数总量表为0.878,各因子α为0.685~0.845;总量表重测信度为0.885,4个因子重测信度为0.693~0.836。结论本研究编制的问卷具有较好的信度和效度,可用于评估大学生心理咨询倾向性。