randomly choosing a sub-group of items

[Marie Sofie]

I would like to generate a random 10 variables from a set of 42 variables and I wonder how i could do it in stata 14.2. Basically, we asked 42 questions measuring similar attitudes and would like to use a random sample of it, otherwise we would have to run many analysis. They do not fit well in factors and the alpha is quite low. Therefore we want to use only a subgroup of it. Is it possible to do so in Stata? Would you please advise a command for it? Thanks a lot.



----------------------- copy starting from the next line -----------------------

* Example generated by -dataex-. To install: ssc install dataex
clear
input byte(PostCons28 PostCons31)
3 2
3 3
5 1
4 2
1 1
1 1
5 1
end

------------------ copy up to and including the previous line ------------------

[Maarten Buis]

How is choosing a random set of items going to help you? That just sounds like it is going to make a bad situation worse. Have you looked at the correlation matrix? Often that is enough to identify obvious clusters or variables that don't belong. Why do you think your variables don't fit in a factor analysis, but at the same you think alpha is suitable?