You want 20 indicator or dummy variables. Note first that it's much easier to use forval
to loop 1(1)20, e.g.
forval j = 1/20 {
gen hou_as_`j' = 0
}
initialises 20 such variables as 0.
I think it's easier to loop over the words of your answer variables, words being here just whatever is separated by spaces. There are at most 20 words, and it is a little crude but likely to be fast enough to go
forval j = 1/20 {
forval k = 1/20 {
replace hou_as_`j' = 1 if word(hou_as, `k') == "`j'"
}
}
Let's put that together and try it out on your example:
clear
input str42 hou_as
"1 2 3 5 11 17 20"
"3 4 8 9 11 14 15 18 20"
"1 3 9 11"
end
forval j = 1/20 {
gen hou_as_`j' = 0
forval k = 1/20 {
replace hou_as_`j' = 1 if word(hou_as, `k') == "`j'"
}
}
Just to show that it worked:
. list in 3
+----------------------------------------------------------------------------+
3. | hou_as | hou_as_1 | hou_as_2 | hou_as_3 | hou_as_4 | hou_as_5 | hou_as_6 |
| 1 3 9 11 | 1 | 0 | 1 | 0 | 0 | 0 |
|----------+----------+----------+----------+----------+----------+----------|
| hou_as_7 | hou_as_8 | hou_as_9 | hou_a~10 | hou_a~11 | hou_a~12 | hou_a~13 |
| 0 | 0 | 1 | 0 | 1 | 0 | 0 |
|----------+----------+----------+----------+----------+----------+----------|
| hou_a~14 | hou_a~15 | hou_a~16 | hou_a~17 | hou_a~18 | hou_a~19 | hou_a~20 |
| 0 | 0 | 0 | 0 | 0 | 0 | 0 |
+----------------------------------------------------------------------------+
Incidentally, your line
local P : subinstr local p "-" ""
does nothing useful. The local macro p
only ever has contents which are integer digits, so there is no punctuation at all to remove.
See also this explanation and
. search multiple responses, sj
Search of official help files, FAQs, Examples, SJs, and STBs
SJ-5-1 st0082 . . . . . . . . . . . . . . . Tabulation of multiple responses
(help _mrsvmat, mrgraph, mrtab if installed) . . . . . . . . B. Jann
Q1/05 SJ 5(1):92--122
introduces new commands for the computation of one- and
two-way tables of multiple responses
SJ-3-1 pr0008 Speaking Stata: On structure & shape: the case of mult. resp.
. . . . . . . . . . . . . . . . . . . . . . . . N. J. Cox & U. Kohler
Q1/03 SJ 3(1):81--99 (no commands)
discussion of data manipulations for multiple response data