st: Snowball sampling

["Ray Hawkins" <sylviantas@gmail.com>]

Hello all,

I am working with social network data, but it is too big, so I would like to
do snowball sampling. My data look like the following. Can you help me
figure out how to keep 'giveid' and corresponding 'recid' for a seed id? For
example, giveid=17 is a seed id. So, I would like to keep giveid=17 and all
giveid (=recid for giveid=17) = 6, 2, 5, 23, 1, 11, 4, 16, 33, 27, 16 (if
exist, of course). Then, for another seed id=32, for example, I would like
to repeat the same process to get certain data size. Thank you in advance.

Ray.


giveid	 recid  n
3	15	1
6	10	1
11	18	3
11	19	3
11	190	3
16	187	1
17	6	11
17	2	11
17	5	11
17	23	11
17	1	11
17	11	11
17	4	11
17	16	11
17	33	11
17	27	11
17	16	11
27	78	2
27	782	2
32	37	7
32	68	7
32	33	7
32	48	7
32	21	7
32	23	7
32	17	7
33	111	1
40	70	5
40	16	5
40	92        5
40	20	5
40	17	5
41	207	6
41	22	6
41	15	6
41	27	6
41	23	6
41	97	6

.........


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/