多水平线性模型在量表问卷调查员间信度评价中的应用

生存质量量表测评的主要内容是调查对象的主观满意度,由于调查员和调查对象双方复杂的心理活动,对于同一问题调查员的提问方式不同,得到的回答可能是不一样的,这种人为因素引起的结果变异称为“调查员间变异”.调查员间变异较小,说明调查员间信度较好,反之亦然.目前评价调查员间信度的常规方法存在的主要问题有二：一是需要重复测量,人为增加了调查成本;二是忽略了调查员和调查对象本身的特征对测评结果的影响.     

两水平方差成份模型在评价调查者间变异中的应用

探讨两水平方差成份模型在评价调查者间变异的应用价值。方法针对问题的应答为连续性正态分布的数据,分别拟合了两水平方差成份模进行对比分析。结果：问卷中各个问题存在着一定的调查者间变异,但逐年改进。结论对于问卷项目的应答为连续性正态分布的数据,在拟合重复调查的情形下,可以采用两水平方差成份模评价问卷项目应答的信度。     

师大学生艾滋病知识及信念的调查分析

了解师大学生艾滋病（AIDS）知识水平、信念及态度，对艾滋病的宣传教育十分重要。为此，我们于1996年11月对山东师范大学在校学生进行了抽样调查，现将结果报告如下。1对象与方法1．1对象：山东师范大学1－4年级学生，每年级随机抽两个班共285人，获得有效答卷280份，应答率98．2％，其中男生124名，女生156名，年龄在18－25岁。1．2方法：自行设计调查问卷，问卷内容包括AIDS的性质、流行与传播、治疗与预防等常识性问题共12题。同时对知识来源及对AIDS的信念和态度也进行了调查，共10题。调查以班为单位。调查员发放已打印好的问卷…     

健康状况自评问卷信度和效度分析

[目的]评价基于世界卫生组织的世界健康调查（WHS）问卷和儿童健康问卷（CHQ）的健康状况自评问卷信度和效度,为开展上海市人群健康水平综合评价奠定技术基础。[方法]在上海市城乡结合部一个居民户数大于400户的居民小区中通过随机抽样获得含有5～15岁儿童的上海户籍家庭80户（371人）作为调查对象。调查问卷以WHS问卷为基础,加入儿童健康问卷一家长卷28（CHQ-PF28）中的相关内容;被调查对象在经培训的调查员指导下完成其健康状况自评。应用重测相关系数和Cronbach α系数评价问卷信度;应用个体功能健康状况评分和总体健康状况的单因素方差评价区分效度;应用主成分法的因子分析评价问卷结构效度。[结果]反映个体健康状况的各维度评分重测相关系数大于0．5,大部分同一维度下各条目的Cronbach α系数均大于0．6,信度较好。自报患病或不良状态的评分明显高于未报患病或不良状态者（P〈0．00）,区分效度较好。测量成人健康条目反映出8个载荷较大的因子,测量儿童健康条目反映出5个载荷较大的因子,与问卷设计接近。[结论]基于WHS和CHQ问卷的健康状况自评问卷可应用于5岁及以上人群健康状况的综合评价。     

《大学生预防艾滋病知识与能力》调查问卷信度与效度分析

目的评价自行设计《大学生预防艾滋病知识与能力》调查问卷的信度和效度。方法随机选择天津外国语学院205名大学生作为调查对象,并对同一人群进行2次问卷调查,计算该问卷的内部一致性、重测信度和结构效度。结果该问卷的6个模块中,大部分模块Cronbach’s系数>0.6,70%的题目重测一致率在0.7或Spearman等级相关系数>0.4。结构效度较好,11个公因子可解释问卷全部内容的60%。结论该问卷的信度和效度较好,可用于评价大学生预防艾滋病知识与能力水平。     

工作内容量表在建筑行业职工中应用的信度和效度

[目的]评估中文版工作内容量表（JobContentQuestionnaire,JCQ）在建筑行业职工中应用的效度和信度。[方法]于2008年4～7月间,采用多阶段随机抽样法抽取包头市在建建筑工地的551名工龄在1年以上的在职职工为研究对象,应用JCQ问卷进行职业紧张状况的调查和评价。[结果]问卷的应答率为95％。信度：JCQ5个模块的同质信度Cronbaeh’sα系数为0．60~0．82。效度：主成分分析方差最大正交旋转提取4个因子,4个因子的条目组成与JCQ量表4个模块组成一致。Pearson相关分析提示问卷主要模块与生理、心理健康状况均具有相关性（P〈0．05或P〈0．01）。[结论]工作内容量表JCQ在包头市建筑行业中使用有较好的信度和效度。     

《肌肉骨骼损伤情况调查问卷》应用于搬运作业人群的信效度评价

目的评价《肌肉骨骼损伤情况调查问卷》在我国搬运作业人群中应用的信度与效度。方法采用整群抽样方法抽取某机场搬运作业人员383人作为调查对象,应用改进后的《肌肉骨骼损伤情况调查问卷》对4个工种搬运作业人员进行调查。计算Cronbach'α系数评价问卷的内部一致性信度,应用结构方程模型进行验证性因子分析评价问卷的结构效度,应用协方差分析检验各个维度得分与不同工种之间的关系并评价问卷的区分效度。结果 (1)内部一致性信度分析,问卷总体Cronbach'α系数为0.808,力量负荷维度和姿势负荷维度的Cronbach'α系数分别为0.559和0.792。(2)结构效度,基于工效学负荷分类构建的理论模型可以与调查数据顺利拟合,修正模型一阶因子的5个因子负荷量分别为0.94、0.95、0.88、0.72和0.80,各一阶因子能够很好的被二阶因子所解释,且两个二阶因子之间存在一定的相关关系。(3)区分效度,协方差分析结果显示不同工种2个维度得分的差异有统计学意义(P0.01)。结论《肌肉骨骼损伤情况调查问卷》具有良好的信度与效度,能够真实有效地反映搬运作业人群职业活动的工效学负荷水平。     

城市体力活动相关建成环境客观评价工具的信度和效度研究

目的 在杭州市6个社区中研究城市体力活动相关建成环境评价工具的信度和效度.方法 在杭州市按公共建筑开发功能区选取6个社区共205个路段,2名调查员同时独立进行调查评价调查者间信度,间隔7d重复调查约一半路段(104个)评价调查者内信度.采用因子分析方法检验结构效度.结果 问卷调查者间信度Kappa值大部分在0.8以上;调查者内信度低于调查者间信度,Kappa值大部分在0.4以上.问卷表面效度良好.结构效度研究提取了6个因子,符合问卷的初始题目设计,因子负荷矩阵显示问卷结构效度良好.结论 此建成环境评价量表的信度、效度较好,适合在杭州使用.     

因子分析在问卷调查中信度效度评价的应用

进行问卷式调查研究的统计分析前，应评价问卷的信度和效度，以确保结果的准确性与可靠性。用单次调查资料评价问卷信度的θ系数法及Ω系数法可克服常用的重复测量法的不足。利用因子分析评价问卷的结构效度是最理想的效度评价方法。本文以“健康行为问卷”为例，介绍因子分析在信度和效度评价方面的应用。分析发现，使用克朗巴哈α系数法低估了整份问卷的内部一致性，而基于因子分析的θ及Ω系数法则较准确。除第２分项外，问卷的５个分项各只有１个公因子，且有较满意的因子载荷量（≥０．４），说明具有良好的结构效度。第２分项的两个公因子可认为分别反映了欲测概念，以及问卷题目的设计方式。     

司法鉴定机构满意度调查表的设计与评价

[目的]编制司法鉴定机构满意度调查表,并评价其在司法鉴定服务质量管理中的应用,为鉴定服务质量的提高提供参考。[方法]选取社会鉴定机构—山西医科大学司法鉴定中心从2008年3～6月期间所接案例中的被鉴定人为调查对象,采用信度分析和因子分析的方法对司法鉴定机构满意度调查表进行评价。[结果]因子分析结果显示,可以将司法鉴定机构工作满意度调查表分为4个维度,第1个维度主要反映鉴定结果的反馈;第2个维度主要反映鉴定服务过程;第3个维度反映鉴定机构的硬件设施;第4个维度反映鉴定人水平。调查问卷的信度分析得克朗巴哈α=0.8251,认为问卷信度较高。[结论]司法鉴定机构满意度调查表信度较高,可以为提高司法鉴定服务质量提供参考依据。     

Importance of Cognitive Testing for Survey Items: An Example From Food Security Questionnaires

Cognitive testing is a useful tool for the development of questionnaire items and can also provide valuable information for the analysis and/or interpretation of preexisting survey data. This paper explains cognitive testing methods and demonstrates the benefits of cognitive testing on existing hunger, food insecurity, and food insufficiency questionnaire items. Semistructured interviews were conducted using retrospective verbal probing techniques. Each interview consisted of two parts: (1) the respondent answered a series of hunger and food insecurity items successively and (2) the interviewer engaged the respondent in a conversation that explored the meaning of the questionnaire items and the respondent's answers to the items. Selected results are described to illustrate how the assessment of questionnaire items is facilitated by cognitive testing. Among other benefits, cognitive testing can clarify the meaning of specific words, determine the best item form and response categories, and help determine question meaning after a survey has been fielded. These results demonstrate that cognitive testing can be beneficial when assessing the quality of questionnaire items and determining item meaning.     

Nonverbal Response Cards Reduce Socially Desirable Reporting of Violence Among Adolescents in Rural Burkina Faso: A Randomized Controlled Trial

PurposeAccurate measures of violence are difficult to obtain from self-reported data because of stigmatization and social undesirability of the topic. Most methods that attempt to reduce such biases require literacy and either remove the benefits of interviewer guidance or do not give individual-level results. We tested a low-tech nonverbal response card that avoids revealing interviewees' responses to interviewers while retaining interviewer support among adolescents in communities with very low educational attainment.MethodsAs part of a broader health questionnaire, we asked a sample of 1,644 adolescents, aged 12–20 years, in northwestern Burkina Faso about their experiences of physical and sexual violence. We randomized participants to either a conventional verbal response arm or a nonverbal response card arm where respondents' answers were unspoken and not displayed to interviewers. We first evaluated response validity and reliability in each arm, then compared prevalence rates across arms and evaluated whether any differences varied by respondent characteristics using regression models.ResultsThe level of internal reliability of responses among nonverbal respondents was similar to or greater than that of verbal respondents. Nonverbal respondents reported similar patterns of physical assault and sexual debut as verbal respondents but significantly higher levels of sexual assault and forced sex. These differences were broadly similar across sample subgroups defined by age, gender, proneness to social desirability, and mental health.ConclusionsNonverbal response cards offer a practical and beneficial method for reducing underreporting of stigmatized and traumatic experiences while maintaining data quality in low-literacy populations.     

Female specialists were more likely to respond to a postal questionnaire about drug treatments for Alzheimer disease

OBJECTIVE: To assess demographic predictors of response for specialists who were mailed a postal questionnaire on drug treatments for Alzheimer disease. STUDY DESIGN AND SETTING: The questionnaire was sent to 317 specialists in Quebec, Canada. Demographic variables included specialty, urban/rural practice, language, sex, and 'number of years since receiving a medical license.' The specialists were stratified according to responder status (i.e., respondent/nonrespondent), and respondents were further stratified as early or late responders. Variables differing between these strata were entered into logistic regression models to see if they predicted response. RESULTS: Only 'female sex' was a predictor of response in the respondent/nonrespondent analysis (OR 2.03; 95% CI 1.17, 3.53). No demographic variables predicted early or late response. CONCLUSION: Researchers planning postal questionnaires should target male specialists with modified or additional mailings to increase response and reduce the potential for nonresponse bias. Caution should be exercised when comparing early vs. late responders as a means of assessing nonresponse bias.     

临床试验中多个终点变量同时评价的多元logistic模型

目的研究临床试验中多个终点变量的同时分析。方法采用多元logistic回归模型,通过对原始资料的格式作适当变换,构造一个虚拟水平,视结果变量为1水平上的观察单位,以患者作为2水平单位,建立2水平logistic模型,对试验组和对照组的疗效,以及患者的年龄,性别,观察指标的基线值,中心效应等协变量进行分析。结果多终点的多元logistic回归模型既可以对单个结果变量进行分析,还可以对多个结果变量进行同时分析,并在扣除组间差异、协变量的影响后,估计两个结果变量之间的相关性。当受试者的多个结果有部分缺失时,该估计方法仍然是有效的。结论多元logistic回归模型可以对多个终点变量进行同时分析。     

Test-retest reliability of the Upper Extremity Questionnaire among keyboard operators.

BACKGROUND: Questionnaires are often used in research among workers although few have been tested in the working population. The Upper Extremity Questionnaire is a self-administered questionnaire designed for epidemiological studies and tested among workers. This study assessed reliability of the instrument. METHODS: A two-part assessment was conducted among 138 keyboard operators as part of a large medical survey. Test-retest reliability was analyzed using the kappa statistic, paired t-test, and intraclass correlation coefficient (ICC). Logistic regression models were used to test the effect of demographic and work-related factors on reliability. RESULTS: The average respondent was a white woman, age 35 years, with some college education, in permanent employment with tenure of 1.4 years. Overall, reports of symptoms were stable from Round 1 to 2. Most kappa values for symptom reports were between 0.60 and 0.89. Kappa values for right and left hand diagrams were 0.57 and 0.28, respectively. Among psychosocial items, Perceived Stress and Job Dissatisfaction Scales were most reliable (ICC = 0.88); co-worker support was least reliable (ICC = 0.44). CONCLUSION: Reliability of items on the Upper Extremity Questionnaire were generally good to excellent. Reports of symptom severity and interference with work were less stable. Demographic and work-related factors were not statistically significant in modeling the variation in reliability. Repeated use of the questionnaire with similar results suggests findings are applicable to a larger working population.     

Interviewer effects in psychiatric epidemiology: a study of medical and lay interviewers and their impact on reported symptoms.

Utilizing data from a field study of psychiatric disorder, the study examined sex differences in patterns of response to interviewers in contrasting status positions. The dependent variables in the analysis were mean scores on nosological scales measuring three dimensions of symptomotology identified in a factor analysis. Independent variables were interviewer status and respondent sex. Three categories of interviewers were randomly assigned to administer a structured interview schedule. Strong interaction effects between interviewer status and respondent sex were noted. Women disclosed significantly fewer symptoms to the high status interviewers (identified physicians), whereas males reported most symptoms to this group. Theoretical explanations derived from the literature on sex roles were offered regarding the origins of women's response to high status figures.     

Comparability of epidemiological information between self- and interviewer-administered questionnaires

The aim of the present study was to determine the extent to which respondents provided the same answers to a health-related lifestyle questionnaire in self- and interviewer-administered forms. A total of 234 subjects completed a 110-item questionnaire in both interviewer and self-administered forms. Modes of administration were separated by a 2-week interval. The order was determined by random allocation. The presence and the extent of the tendency to give socially acceptable responses were evaluated using percentage of bias calculated as the ratio of the difference in proportion of positive responses or the mean between interviews and questionnaires and those in questionnaires. All percentages of bias were in the positive direction, ranging from 1.4% (physical exercise) to 26.1% (general life stress). The average percentage of bias was higher in women than in men and were stronger for younger respondents. The age differences between interviewer and respondent were inversely and most strongly related to percentage of bias. Self- and interviewer-administered questionnaires yield very similar results in discriminating between subjects, but the interviewer-administered questionnaire showed systematically more desirable responses to questions related to lifestyle factors. The differences in characteristics between interviewer and respondent may be important determinants of the socially desirability bias in the interview.     

A qualitative validation of the Minnesota Living with Heart Failure Questionnaire

BACKGROUND: In an overview of quality-of-life questionnaires that have been used to measure treatment effects in clinical trials in congestive heart failure (CHF), it was concluded that 'no QoL instrument has measured quality of life in CHF trials in a reliable or reproducible fashion' and that the usefulness of these instruments in clinical trials was hampered by a lack of sensitivity to changes in health status. AIM: The aim of our study was to qualitatively explore issues of validity, specificity, and sensitivity with regard to the Minnesota Living with Heart Failure (MLHF) questionnaire in patients with severe heart failure. METHODS: We included 31 patients with severe heart failure in a 2-year prospective longitudinal case study. Each patient was interviewed two or three times in a semi-structured way, resulting in a total of 76 interviews. We used simple qualitative techniques such as respondent observation and respondent debriefing to collect data on practical and interpretative problems that patients experience in conducting the task of completing the questionnaire. RESULTS: The introduction to the questionnaire, which contains essential instructions, was almost never read. Patients did not interpret items as intended. Some items posed problems to respondents because of their wording or because they were considered not applicable to the respondents' circumstances. CONCLUSIONS: Our findings suggest that the MLHF does not measure the concept that it is intended to measure, but our study does not provide a remedy. However, problems regarding individual items and response categories can be remedied.     

Factors in errors of omission on a self-administered paper questionnaire

This article examines the role of question and respondent characteristics on omission errors made on the 2007 Health Information National Trends Survey (HINTS) questionnaire. Higher omission error rates were found for items with open-ended response formats, items placed outside of the body of the questionnaire, and items following skip instructions. Respondent and survey completion characteristics seen to impact omission error included age, education level, household income level, and the amount of time respondents reported having spent on the questionnaire.