Multilevel modelling of complex survey data
Multilevel modelling of complex survey data
复制标题
DOI：
10.1111/j.1467-985x.2006.00426.x
复制
发表时间：
2006-01-01
影响因子：
2
通讯作者：
Skrondal, Anders
中科院分区：
文献类型：
--
作者：
Rabe-Hesketh, Sophia;Skrondal, Anders
多水平模型有时用于涉及多阶段抽样、不等抽样概率和分层的复杂调查数据。我们考虑了广义线性混合模型，特别是二分反应的情形。利用自适应求积，实现了一种在任意层数的多层模型中调节逆概率权重的伪线性方法。三明治估计器用于获得考虑分层和聚类的标准误差。当使用在簇中的基本单位之间变化的级别1权重时，权重的缩放变得重要。我们指出，当响应是二分的时，不仅方差分量，而且回归系数都可能严重偏向。使用STATA程序GLLAMM对2000年《国际学生评估计划》研究的美国样本中有关阅读水平的复杂调查数据进行了伪随机性分析，GLLAMM可以估计广泛的多水平和潜在变量模型。在蒙特卡罗实验中，研究了不同方法处理第一级权重的伪极大似然方法的性能。(条件)回归系数的伪最大似然估计器对于大的簇大小表现良好，但对于小的簇大小则有偏差。相比之下，边际效应估计器在这两种情况下都表现良好。我们的结论是，当使用级别1权重时，对于小聚类大小的伪最大似然估计必须谨慎。
Multilevel modelling is sometimes used for data from complex surveys involving multistage sampling, unequal sampling probabilities and stratification. We consider generalized linear mixed models and particularly the case of dichotomous responses. A pseudolikelihood approach for accommodating inverse probability weights in multilevel models with an arbitrary number of levels is implemented by using adaptive quadrature. A sandwich estimator is used to obtain standard errors that account for stratification and clustering. When level 1 weights are used that vary between elementary units in clusters, the scaling of the weights becomes important. We point out that not only variance components but also regression coefficients can be severely biased when the response is dichotomous. The pseudolikelihood methodology is applied to complex survey data on reading proficiency from the American sample of the 'Program for international student assessment' 2000 study, using the Stata program gllamm which can estimate a wide range of multilevel and latent variable models. Performance of pseudo-maximum-likelihood with different methods for handling level 1 weights is investigated in a Monte Carlo experiment. Pseudo-maximum-likelihood estimators of (conditional) regression coefficients perform well for large cluster sizes but are biased for small cluster sizes. In contrast, estimators of marginal effects perform well in both situations. We conclude that caution must be exercised in pseudo-maximum-likelihood estimation for small cluster sizes when level 1 weights are used.