Random dummy variable

[Catherine Vassen]

I had to setup a new dataset with 100 observations. Now, what is being asked is to create a dummy called prep for students who have taken a similar class before. The dummy should
be equal to one for 25 of the students. I am a student, and we are struggling on this one. we'd be more than grateful for any help.
Best,
Catherine

[Nick Cox]

This is an FAQ: http://www.stata.com/support/faqs/st...les/index.html

Just shuffle them randomly then pick off the first (or the last) 25.

clear
set obs 100
set seed 2803
gen id = _n
gen random = runiform()
sort random
gen wanted = _n <= 25

Note that using a condition like runiform() < 0.25 won't ensure exactly 25/100, but this technique will.

[Mathias Pedersen Heinze]

Hi Catherine,

I'm not exactly sure what you are asking for? How do you know which students have attended the class before? Or do you just want a variable called prep with 25 values of 1 and 75 values of 0? If the latter is the case, you can generate it like this:

clear
set obs 100

gen prep = inrange(_n, 1, 25)

[Sebastian Geiger]

Catherine,

This should work:

gen random = runiform()
sort random
gen dummy = 0
replace dummy = 1 in 1/25

Edit: Ok, Nick was faster ;-).

[Catherine Vassen]

thanks a lot guys! :-)