To find the variables we need to answer this question, we could either use the BAIS questionnaire or use the following lookfor commands:
lookfor hiv
lookfor test
After scanning the list of variables, we see that the variables hivtest3 (would you ever want to be tested for hiv) and hivplace (do you know of place to get an hiv test) are the variables we need to answer this question. Using these variables we can create a bivariate table:
tab hivtest3 hivplace
would you |
ever want to |
be tested | do you know of place
(again) for | to get an hiv test
hiv | Yes No | Total
-------------+----------------------+----------
Yes | 1378 635 | 2013
No | 703 577 | 1280
Undecided | 240 135 | 375
-------------+----------------------+----------
Total | 2321 1347 | 3668
Now we are getting close to answering the question. For the final step, we must restrict our analysis to respondents that are over the age of 20. The table above gives us the distribution for the the entire sample, so we need to use a qualifier to restrict the table to respondents who are 21 and older.
tab hivtest3 hivplace if age > 20 & age ~=.
would you |
ever want to |
be tested | do you know of place
(again) for | to get an hiv test
hiv | Yes No | Total
-------------+----------------------+----------
Yes | 957 326 | 1283
No | 473 275 | 748
Undecided | 178 70 | 248
-------------+----------------------+----------
Total | 1608 671 | 2279
Looking at the table above, we see that of the respondents who are over the age of 20, 957 of them would want to be tested for HIV and know of a place to get tested. It would be interesting to see what other variables interact with the bivariate relationship above (possible project idea).