how to get summary statistics across groups rather than within group?

[Lucrecia Lei]

Hi everyone,

I'm using Stata to deal with a hierarchical dataset, where I have level one data (individual) and level two data (household). My data looks like this:

ind_id    household    income    housesize
-----------------------------------------------
1            1            12000        57
2            1            39000        57
3            2            27560        140
4            2            39000        140
5            2            25000        140
6            3            27435        120
7            3            22390        120
-----------------------------------------------

So here each individual from the same household has same housesize. I want to get the mean and standard deviation of housesize, for instance with the data here it should be (57+140+120)/3 Which command should I use?

And since I have many variables like housesize, and I want mean, sd, max of all of them in one file, is there any command that do this directly?


Thank you so much!

[Andrew Musau]

egen tag= tag(household)
*STATISTICS
sum housesize if tag
*VARIABLES
egen wanted1= mean(cond(tag, housesize, .))
egen wanted2= sd(cond(tag, housesize, .))

[Lucrecia Lei]

egen tag= tag(household)
*STATISTICS
sum housesize if tag
*VARIABLES
egen wanted1= mean(cond(tag, housesize, .))
egen wanted2= sd(cond(tag, housesize, .))

Thank you Andrew! This is exactly what I want!

[Lucrecia Lei]

And just figure out another way is to use collapse, but restore is annoying to me, tag is much better!