Try -mrtab-. I don't know how well it performs if you have "hundrets" of codes (-mrtab- internally generates a dummy for each code and, therefore, might be quite memory consuming in your case). Please let me know if you encounter any problems, and I'll see what I can do about it. To install -mrtab-, type
. ssc install mrtab
ben
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of Americo
> Sent: Saturday, March 20, 2004 8:20 PM
> To: [email protected]
> Subject: st: Multiple Response Analisys
>
>
> Hi, colisters.
>
> I have a problem on analizing a multiple response variable of a
> Domiciliary Survey.
>
> It consists in a variable with hundreds of possible responses, and
> each one of them assumes an alfanumeric code. Now, I have the same
> variables in numeric codes, if it helps for something.
>
> The big problem is that there were some subjects that
> answered more than one response up to the number of 6 responses for
> the same variable.
>
> So, now I have a database with 6 variables for the same question in 6
> diferent colums (J02, J03, J04, J05, J06 and J07).
>
> If there was just a few possible codes, I could make a few
> dummies for each
> code and analize them separately, but as there are hundreds of
> them, it is impossible to me to analize them as dummies
> variables for each
> possible code.
>
> When I try to get the frequencies of those codes I only know
> how to get
> the frequencies of each colum, using the svy commands, but have no
> idea on how to have all of them analized at the same time.
>
> As I have each one of the codes listed on any of the 6 colums
> depending on the
> order of response, I will lose data if I analise each columm
> separated.
>
> In SPSS I know what I could do to avoid this problem. I would
> make a Multiple Response Set,
> grouping all the colums in a set that would be analized at the same
> time to calculate the frequencies. But I have no idea on how to do
> the same thing with Stata.
>
> If someone know how to group sets of variables to be
> analized together, please help me.
>
> Thanks on advance,
>
> Americo F. Pelicioni
> Epidemiology Professor
> S�o Paulo - Brazil
>
>
> *
> * For searches and help try:
> * http://www.stata.com/support/faqs/res/findit.html
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
