Design Effect Caveats

被引：3

作者：

Gambino, Jack G. ^{[1
]}

机构：

[1] STAT Canada, Household Survey Methods Div, Ottawa, ON K1A 0T6, Canada

来源：

AMERICAN STATISTICIAN | 2009年 / 63卷 / 02期

关键词：

Clustered design; Complex survey; Sample design; Variance estimation;

D O I：

10.1198/tast.2009.0028

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

Design effects typically are used either to compare the efficiency of survey designs or to make it possible to use simple formulas to compute the variance of estimates from complex surveys. Here we discuss some of the pitfalls that may be encountered in practice when a design effect is computed incorrectly or the wrong design effect is used.

引用

页码：141 / 146

页数：6