Dropping lowest n values of a group variable

[Kartikeya Bhatotia]

Hi, my question is a bit silly but I am just not being very bright today

As you can see from the dataset below, I have arranged scores in a descending order. I want a shortlist that has more women than men, so let's say I want to drop lowest n number of observations but they should only be men - how do I do it? Here women are 1 and men are 0

* Example generated by -dataex-. To install: ssc install dataex
clear
input float(scgender score)
1 10
1 9.5
1 9.5
1 8.5
1 8.5
1 8.5
1 8.5
1 8.25
1 8
0 8
0 8
0 8
0 8
1 8
1 8
0 7.75
1 7.75
1 7
1 7
1 7
1 7
1 7
1 7
1 7
1 7
1 7
1 7
1 7
1 7
1 7
1 6.75
1 6.75
1 6.75
1 6.75
1 6.75
1 6.75
1 6.75
1 6.75
1 6.75
1 6.75
1 6.75
0 6.5
0 6.5
0 6.5
0 6.5
1 6.5
1 6.5
0 6.5
0 6.5
0 6.5
1 6.5
1 6.5
0 6.5
1 6.5
0 6.5
0 6.5
0 6.5
0 6.5
0 6.5
1 6.5
0 6.5
0 6.5
0 6.5
0 6.5
0 6.5
0 6.5
0 6.5
0 6.5
1 6.5
0 6.5
0 6.5
1 6.5
0 6.5
0 6.5
0 6.5
0 6.5
0 6.5
1 6.5
0 6.5
0 6.5
0 6.5
1 6.5
0 6.5
0 6.5
1 6.5
0 6.5
0 6.5
0 6.25
1 6.25
1 6.25
0 6.25
1 6.25
1 6.25
1 6.25
1 6.25
0 6.25
0 6.25
0 6.25
1 6.25
0 6.25
end

[Carlo Lazzaro]

Kartikeya:
assuming that your cut-off for male score is, say, 6.5:

drop if score<=6.5 & scgender==0

[Nick Cox]

You have a major problems with ties there.

[Joro Kolev]

If women are 1, and men are 0s (from heroes to zeroes, oh what a sad faith), and then n=50, then you can try

. gsort -scgender -score

. drop in 50/l
(51 observations deleted)

In the above example of n=50 this procedure drops all men and then some of the women with lowest scores.