if you administered a census questionnaire to an entire group of interest ("population") and got a 68% response rate. Is it wrong to calculate confidence intervals for the medians of the data subsets? The data subsets contain outliers and medians were calculated. Tx!
Asked
Active
Viewed 28 times
0
-
It seems that you are you are using two unregistered accounts. Please, follow the [instructions on our help page](http://stats.stackexchange.com/help/user-merge) to merge them and don't forget to register your account once and for all. – chl Mar 11 '13 at 11:57
-
I've marked this as a possible duplicate (as well as a certain duplicate of another, now closed question). The thing that might make it different is that you are now asking specifically about medians. If this is a question about medians I think it is ok to stand; if it is a repeat of the more general question about confidence intervals it is a duplicate. However, I can't think of any way it would be wrong to calculate confidence intervals for the medians other than because of the missing data problems discussed in the possible duplicate. – Peter Ellis Mar 11 '13 at 18:24