整群抽样调查数据分析中应正确计算抽样误差

为了澄清整群抽样调查数据分析中正确计算抽样误差的必要性，以在某市15岁及以上人群中开展的一次两阶段整群抽样调查为例，分别采用适用于单纯随机抽样数据的方法和考虑了复杂抽样设计的方法对数据进行分析。结果显示，忽略对复杂抽样设计的考虑，不恰当的采用适用于单纯随机抽样数据的方法进行数据分析，不仅有可能大大低估样本统计量的抽样误差，在进行假设检验时，甚至会得到错误的结果，故正确分析和报告整群抽样调查数据的抽样误差是非常必要的。     

多级整群抽样样本大小的简便估计

多级整群抽样样本大小的简便估计合肥联合大学计算机工程系黄体乾整群抽样是将总体分成Ｎ个次总体即Ｎ个群或单位或单位群，直接从总体划分出的群称一级群。然后从Ｎ个一级群中随机抽取ｎ群，记ｆ１＝ｎ／Ｎ，ｆ１为一级单位群抽样比例。如果对这ｎ个一级群均作为个体调查...     

整群抽样所需样本大小的简便估计

整群抽样所需样本大小的简便估计合肥联合大学黄体乾安徽医科大学施仲赋合肥农经学院韩嵘抽样调查是从调查总体中抽取一定数量的单位组成样本，然后对样本中的单位进行调查。在大规模抽样调查中，整群抽样是经常采用的方法。整群抽样是先将总体分成Ｎ个群，然后从Ｎ个群中...     

国家人体生物监测项目抽样方法及误差估计

目的 探讨国家人体生物监测项目抽样方法,并通过计算抽样误差来评估抽样设计。方法 该监测采用多阶段复杂抽样方法,并以贵州省抽样结果为例,应用多阶段不等概率抽样误差计算方法计算贵州省抽样误差及变异系数,探讨多阶段抽样设计下误差估计方法。结果 该监测覆盖全国31个省（自治区、直辖市）的152个监测点,抽取21 888名3~79岁城乡居民,三个阶段置换率分别为5.26%、6.35%及40.6%。经计算贵州省抽样误差为3 207 594人,变异系数为0.097。结论 根据多阶段不等概率抽样方法计算,贵州省抽样变异系数较小,精度较高,但该计算方法未考虑人群缺失率、应答率等非抽样误差的权重调整,大型公共卫生监测项目多阶段抽样误差计算方法有待继续研究。     

敏感问题Simmons模型的(分层)整群抽样研究

目的 为敏感性问题提供科学的较复杂抽样调查方法 及其统计量的计算公式.方法 Cochran W.G.的抽样理论及给出的抽样研究结果 ,随机应答技术的Simmons模型,全概率公式,方差的基本性质等理论与方法 被利用.结果 推导出二分类敏感问题随机应答技术Simmons模型在整群抽样、分层整群抽样下总体比例的估计量及其估计方差的计算公式,并对相应的调查方法 与技术进行了科学的设计.结论 本文提供的敏感问题Simmons模型下整群抽样、分层整群抽样的方法 信度较高,值得推广应用.     

县乡级计划免疫接种率调查的抽样方法

目的：研究县乡级计划免疫接种率调查的抽样方法。方法：在充分利用计划免疫以往信息的基础上，采用整群抽样与二阶段整群抽样方法来实现。结果：建立的方法经计算机模拟与现场验证令人满意。结论：在９５％可信限与±１０％误差条件下，通常对乡级抽取２～３个行政村，对县级抽取６～８个行政村调查，即可反映计划免疫接种率状态。     

以乡为单位儿童免疫接种率调查的抽样方法研究

在充分利用计划免疫已往信息的基础上,分别采用整群抽样方法与二阶段随机整群抽样方法,进行以乡、以县为单位的接种率调查,在95％可信限与±10％误差条件下,对乡级抽取2～3个行政村,对县级抽取6～8个行政村,即可达到调查目的,用最小的样本容量来反映出客观接种率状态,经现场验证与计算机模拟,结果令人满意．     

敏感问题双无关问题模型分层二阶段整群抽样的统计方法及应用

目的为二分类敏感性问题提供科学的较复杂抽样调查方法及其统计量的计算公式。方法采用Coch-ranW.G.的抽样理论及研究结果、随机应答技术的双无关问题模型、分层两阶段整群抽样调查方法、全概率公式、方差的基本性质、效度评价等理论和方法。结果推导出二分类敏感问题双无关问题模型在分层二阶段整群抽样下总体比例的估计量及其方差的计算公式,并成功应用该调查方法及相关公式调查估计得苏州大学新校区学生婚前性行为发生比例为16.77%,将其与效度准则指标作假设检验得P=0.1696。结论本文提供的敏感问题双无关问题模型的分层二阶段整群抽样调查方法及相应的统计量计算公式科学有效。     

Monte Carlo抽样计算配对符号秩检验的确切概率

配对符号秩检验法(Wilcoxon matched-pairs signed-ranks test)是一种处理配对资料的非参数检验方法.由于其确切概率的经典算法[1,2]计算量非常大,所以传统的配对符号秩检验常常使用查表法或者正态近似法.但是,常用的医用统计书中的配对符号秩检验的界值表并不完全一致[1,2].而在样本量不够大时,正态近似效果不好.所以需要一种实际可行的确切概率的计算方法.     

数量特征敏感问题的整群抽样调查方法研究及应用

目的寻求在整群抽样下,数量特征敏感问题的科学调查方法及相应统计计算公式。方法采用数量特征敏感问题无关联、加法、乘法三种随机应答技术(RRT),整群抽样调查方法,均值的性质、经典的抽样理论等概率统计理论和方法,重测信度、和谐信度评价方法。结果分别推导出三种RRT在整群抽样下,总体均值估计量及其方差的计算公式;成功应用这三种调查方法及相关公式调查得苏州大学近两个学期学生考试作弊平均次数分别为1.1050、1.0375、0.9964,两次重测信度、三种RRT和谐信度的相关系数均大于0.94,P<0.01。结论数量特征敏感问题三种RRT整群抽样调查方法及相应统计计算公式科学可行。     

Empirical modelling of population sampling: lessons for designing sentinel surveillance

BACKGROUND: In order to consider the practical viability of 1% sentinel area surveillance for health information in resource-poor settings where complete registration is unrealistic, the effects of different sampling procedures on the representativeness of 1% population samples have been investigated. METHODS: Using the 1991 census of England as a basis for modelling, 20 1% samples, each incorporating seven key parameters, were drawn at random from the overall dataset by each of eight different sampling procedures. Each sample was compared with the 'gold standard' of the overall census results, enabling comparisons between the different sampling procedures. RESULTS: Representativeness of the 1% samples varied considerably between parameters and sampling procedures. At one extreme, the proportion of males in the population was distributed such that different sampling methods had little effect. On the other hand, samples of a heterogeneous parameter such as the proportion of non-whites in the population depended greatly on the procedure used. Sampling smaller administrative units tended to be more accurate. However, sampling units using probability proportional to size generally gave less representative samples. Stratifying urban and rural populations in the samples had little effect. Multistage sampling, emulating typical demographic surveillance sites, also generally gave less representative samples. CONCLUSIONS: It is possible to achieve representative data by taking 1% of a national population in a sentinel surveillance approach, but sampling design can have an important influence on the outcome. This modelling supports the concept of 1% sentinel surveillance for health information in poorer settings, where complete data are unavailable.     

复杂抽样调查数据实例分析

提出复杂抽样调查数据的分析思路和方法以及忽视权重和抽样设计时会出现的问题.文中以2002年中国居民营养与健康状况调查数据中高血压患病率的估算为例,分加权和不加权、考虑和不考虑整群设计特征的四种组合情况对数据进行分析.表明忽视权重的设置会影响点估计和标准误的估计,忽视对整群设计特征的考虑不仅会高估结果的精确度,还会得到地区间患病率有差异的假阳性结论.因此使用合理的统计方法分析复杂抽样调查数据非常有必要.     

Extending cluster lot quality assurance sampling designs for surveillance programs

Lot quality assurance sampling (LQAS) has a long history of applications in industrial quality control. LQAS is frequently used for rapid surveillance in global health settings, with areas classified as poor or acceptable performance on the basis of the binary classification of an indicator. Historically, LQAS surveys have relied on simple random samples from the population; however, implementing two‐stage cluster designs for surveillance sampling is often more cost‐effective than simple random sampling. By applying survey sampling results to the binary classification procedure, we develop a simple and flexible nonparametric procedure to incorporate clustering effects into the LQAS sample design to appropriately inflate the sample size, accommodating finite numbers of clusters in the population when relevant. We use this framework to then discuss principled selection of survey design parameters in longitudinal surveillance programs. We apply this framework to design surveys to detect rises in malnutrition prevalence in nutrition surveillance programs in Kenya and South Sudan, accounting for clustering within villages. By combining historical information with data from previous surveys, we design surveys to detect spikes in the childhood malnutrition rate. Copyright © 2014 John Wiley & Sons, Ltd.     

Single-stage cluster sampling with a telescopic respondent rule: a variation motivated by a survey of dementia in elderly residents of Shanghai

In this report, we consider the situation in which one wishes to identify a cohort of a specified number of individuals within each of several domains for future follow-up studies based on a single-stage cluster sampling design. We develop sample size formulae relevant to this situation and introduce a variation of single-stage cluster sampling that seems more suitable in this situation than is ordinary single-stage cluster sampling. The basis for this variation is the concept that the definition of eligible respondents is not the same for all clusters. The use of this modified respondent rule (which we call telescopic) enables one to meet specified sample sizes in all domains of interest without the need to sample extra individuals in some domains. We used a version of this sampling design successfully in the field with a survey of elderly persons conducted in Shanghai, People's Republic of China.     

建立"中国食物与营养监测系统"的抽样方案和质量控制

在卫生部145个疾病监测点与国家统计局住户调查点一致的基础上,分为城市、一般农村和较贫困农村3层,进行抽样。按6岁以下儿童低体重患病率为5％水平估计样本量需抽取160000例,在全国抽取40个点,（城市14个,一般农村17个,较贫困农村9个）分布在全国26个省、市、自治区。每个监测点采取多阶整群抽样方法,抽取400例6岁以下儿童,40个点共抽取160000例。40个监测点的人口共2067．8万人,占全国人数的1．7％。通过对抽取样本的儿童年龄分布与总体进行拟合度检验,检验结果表明,样本与总体分布是一致的。说明样本对总体的代表性较好。     

A temporal-spatial cluster of sudden infant death syndrome in Navarre,Spain

An apparent temporal-spatial cluster of Sudden Infant Death Syndrome (SIDS) was noted in the Autonomous Community of Navarre, Spain, when four unrelated children aged between 1 and 6 months died unexpectedly within a 7-day interval in January, 1990. The population under one year of age in Navarre was approximately 4,800. The scan test of temporal clustering showed that the sudden infant deaths occurred closer to one another in time significantly more often than would be expected by chance. All four infants lived in a neighbourhood of the capital of Navarre, which accounts for approximately half the region's population. The clustered cases coincided with an outbreak of influenza type A detected by the epidemiological surveillance system and seen by the increase in 1990 over the same period in the previous year in the number of paediatric emergency-ward admissions during the epidemic days. The results confirm the presence of a temporal-spatial cluster of SIDS and favour an environmental etiology where exposure to influenza A viruses is implicated.     

北京市2017年同伴推动抽样法招募男男性行为人群抽样效果分析

目的 评价2017年北京市同伴推动抽样（RDS）法招募MSM人群的随机性和代表性。方法 采用RDS法招募北京市MSM人群进行面对面问卷调查和HIV检测。结果 共招募MSM 600人,其中种子14人,种子的网络规模中位数为10人,高于其他被招募的MSM,招募人数按轮次呈正偏态分布,第4轮招募人数最多。不同种子招募人数差异较大,招募链人数最多的为184人、113人和92人,5个种子的招募链人数不足10人,其中以2名大学生种子招募人数最少,仅各招募1人。主要人口学指标在5~9层达到平衡,各指标绘制的收敛图最终趋于平稳,瓶颈图除婚姻状况和文化程度中部分曲线收敛趋势不明显外,其他曲线最终趋于收敛。主要指标的同质系数除文化程度外均在1左右。HIV感染率为7.9%（95% CI：4.4%~11.4%）。结论 采用RDS法招募的MSM样本,主要人口学指标达到平衡,招募过程合理可控,在一定程度上可以代表北京市MSM人群。     

Optimal auxiliary‐covariate‐based two‐phase sampling design for semiparametric efficient estimation of a mean or mean difference,with application to clinical trials

To address the objective in a clinical trial to estimate the mean or mean difference of an expensive endpoint Y, one approach employs a two‐phase sampling design, wherein inexpensive auxiliary variables W predictive of Y are measured in everyone, Y is measured in a random sample, and the semiparametric efficient estimator is applied. This approach is made efficient by specifying the phase two selection probabilities as optimal functions of the auxiliary variables and measurement costs. While this approach is familiar to survey samplers, it apparently has seldom been used in clinical trials, and several novel results practicable for clinical trials are developed. We perform simulations to identify settings where the optimal approach significantly improves efficiency compared to approaches in current practice. We provide proofs and R code. The optimality results are developed to design an HIV vaccine trial, with objective to compare the mean ‘importance‐weighted’ breadth (Y) of the T‐cell response between randomized vaccine groups. The trial collects an auxiliary response (W) highly predictive of Y and measures Y in the optimal subset. We show that the optimal design‐estimation approach can confer anywhere between absent and large efficiency gain (up to 24 % in the examples) compared to the approach with the same efficient estimator but simple random sampling, where greater variability in the cost‐standardized conditional variance of Y given W yields greater efficiency gains. Accurate estimation of E[Y | W] is important for realizing the efficiency gain, which is aided by an ample phase two sample and by using a robust fitting method. Copyright © 2013 John Wiley & Sons, Ltd.     

2004-2005年全国死因回顾抽样调查胃癌死亡率分析

目的 了解日前我国居民胃癌死亡特征及变化趋势.方法 分析2004-2005年全国死因回顾抽样调查中158个全国样本点的胃癌死亡数据,并与前2次全国死因调查结果比较.结果 2004-2005年我国样本地区胃癌粗死亡率为24.71/10万(35 250/142 660 482),中国人口标化死亡率(简称中标率)为16.16/10万,占恶性肿瘤死因构成的18.19%(35 250/193 841),居第3位.本次调查结果与1973-1975年我国样本地区胃痛粗死亡率(17.40/10万)和中标率(17.70/10万)相比分别上升42.01%和降低8.70%,与1990-1992年我国样本地区胃癌粗死亡率(25.16/10万)和中标率(21.76/10万)相比分别降低1.79%和降低25.74%,位次由前2次调查统计的第1位后移至第3位.2004-2005年城市样本地区胃癌粗死亡率为22.98/10万(11 005/47 899 806),中标率为13.63/10万,占肿瘤死冈构成的15.30%(11 005/71 936),居第3位,与1973-1975年城市样本地区胃癌粗死亡率(19.44/10万)和中标率(19.80/10万)相比分别升高18.21%和降低31.16%,与1990-1992年城市样本地区胃癌粗死亡率(19.44/10万)和中标率(15.34/10万)相比分别升高18.21%和降低11.15%;农村样本地区胃癌粗死亡率为25.59/10万(24 245/94 760 676),中标率为17.64/10万,占肿瘤构成的19.89%(24 245/121 905),居第3位,与1973-1975年农村样本地区胃癌粗死亡率(16.62/10万)和中标率(17.00/10万)相比分别升高53.97%和升高3.76%,与1990-1992年农村样本地区胃癌粗死亡率(27.16/10万)和中标率(24.36/10万)相比分别下降5.78%和下降27.59%.结论 目前胃癌仍然是我国重要的恶性肿瘤死因之一.我国样本地区胃癌标化死亡率明显下降,提示我国近期经济社会的发展有益于降低胃癌的发病风险.