e*****8 发帖数: 30 | 1 刚实习遇到的SAS问题:
一个关于survey的data,把参与调查的人分成若干组,然后在每个组都问相同的若干问
题。问题答案用数字表示,每个数字表示不同程度的看法。要求个within group
variance 和 between group variance 以及 ICC ( intraclass correlation). 应该
用到哪些关键SAS code阿? 请问这些variance 应该怎么理解阿? 比较常用的anova
例子是比较不同的treatment 对农作物高度影响。 对于这种survey data, 不知道怎么
用anova分析了。。。望解疑答惑,感激不尽!! | k*******a 发帖数: 772 | 2 survey data一般用proc surveymeans 吧
看下相关的manual应该就清楚了 | h**********1 发帖数: 155 | 3 there are some specific proc in sas to deal with survey data.
also, you can try r package survey and there is a book about this book
written by Thomas Lumely
moreover, you can try sas-callable sudaan
in order to deal with survey data, you also need to know the design, weight
for each subject etc... | p*******i 发帖数: 1181 | 4 proc surveymeans没用过 不过within / between group variance在proc mixed里面设
一个repeated /subject=xxx就有了,ICC就是用那两个算的吧。如果仅仅是把survey的
结果当一个ordinal var似乎这样就可以了。 |
|