- 积分
- 1743
好友
记录
日志
相册
回帖0
主题
分享
精华
威望 旺
钢镚 分
推荐 人
|
National Health and Nutrition Examination Survey: Analytic Guidelines, 2011-2014 and 2015-2016
December 14, 2018中提到:
Various sample weights are available on the data release files. Use of the correct sample weight for NHANES analyses depends on the variables being used. A good rule of thumb is to use “the least common denominator” where the variable of interest that was collected on the smallest number of respondents is the “least common denominator.” The sample weight that applies to that variable is the appropriate one to use for that particular analysis.
Sampled participants who completed the interview and were eligible for the examination, but did not respond, were assigned non-zero interview weights and examination weights of zero. Records with a zero examination weight should be treated as missing when the exam data are analyzed. For example, if all variables come from the interview and exam, then the sample used in the analysis should reflect only those with non-zero exam weights and exam weights should be used in the analysis. Similarly, if any variable used comes from a specific subsample, then the sample used in the analysis should only represent those with a non-zero subsample weight and the subsample weights should be used in the analysis.
也就是说虽然我整合了包括demographic、questionnaire、laboratory的数据,但是因为我是整体分析的,只要加权一个统一有的WTINT2Y或者WTMEC2Y就行了? |
|