有关多个观测者测量结果的一致性检验

yushandour 发表于 2013-6-5 16:23
楼主，你的问题解决了吗？STATA里面倒是可以做多个观察者间的一致性

您好，请问如何在ＳＴＡＴＡ里做多个观察者间的Ｋａｐｐａ一致性，谢谢！！！

yushandour 发表于 2013-6-5 16:23
楼主，你的问题解决了吗？STATA里面倒是可以做多个观察者间的一致性

请问是怎么计算Fleiss’ multirater kappa的？

You can download and run the SAS macro called 'magree' developed by SAS.  

to download: http://support.sas.com/kb/25/addl/fusion_25006_2_magree.sas.txt

let us say you have data like: 4 subjects, 3 raters, and score y
data a;
input s r y @@;
datalines;
1 1 1
1 2 2
1 3 2
2 1 1
2 2 1
2 3 3
3 1 3
3 2 3
3 3 2
4 1 2
4 2 3
4 3 2
;
run ;

then you can run the following codes to get your kappa agreement statistics:

%magree (data=a ,
   items=s,
   raters=r,
   response=y)

if you also want to learn more ICC here is review paper:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3402032/pdf/nihms372951.pdf

https://www.amazon.com/Handbook-Inter-Rater-Reliability-Definitive-Measuring/dp/0970806280