Pritpal,
maybe the following is doing what you want:
gen first=""
gen second=""
gen third=""
foreach var of varlist Brand* {
replace first = "`var'" if `var'==1
replace second = "`var'" if `var'==2
replace third = "`var'" if `var'==3
}
Afterwards you may drop all other variable but first, second, and third.
Regards
Sebastian
On 4/13/07, Pritpal Marjara <[email protected]> wrote:
In a multiple choice question in a population based survey, I have asked for
the top three preferred brand of health services.
The total number of codes are twelve in number to choose from. At the time
of data entry, the data was entered as twelve variables of brand names as
columns. For each case, nine of the variables thus have missing values and
remaining 3 variables have 1,2 or 3 numerical value as a preference in
order. I am wondering if there is any way to transpose these twelve
variables with the values without affecting remaining 300+ variables in the
database.
For example, the data now is;
ID# Brand1 Brand2 Brand3 Brand4 Brand5 Brand6 Brand7 ....
001 - 3 - 1 - 2 -
002 1 - - 2 - 3 -
003 - 1 3 - 2 - -
And, I wish if the data would look like;
ID# first second third
001 brand4 brand6 brand2
002 brand1 brand4 brand6
003 brand2 brand5 brand3
I am using Stata 8 version.
Appreciate guidance from the group in resolving this in a simpler way.
Thank you,
Pritpal
*
* 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/
