Scott's example code underlines that this is indeed a -reshape- problem and that you just need the one trick of creating an identifier that will be used for the columns.
Other trickery in this territory is detailed at
FAQ . . . . . . . . . . . . . . . . . . . . . . . . Problems with reshape
12/03 I am having problems with the reshape command. Can
you give further guidance?
http://www.stata.com/support/faqs/data/reshape3.html
But a bigger question is why you want to do this. On the whole you are better off with your existing data structure. Although working rowwise is possible and often natural, as will be explored in some detail in Stata Journal 9(1) 2009, it is difficult to think of anything easier with your new data structure.
Nick
[email protected]
Scott Merryman
clear
input pid str4 site str9 datestamp bglevel str9 enrolldate str9 fuploaddate
1424 "PBCI" "10jan2006" 253 "09jan2006" "13jan2006"
1424 "PBCI" "10jan2006" 185 "09jan2006" "13jan2006"
1424 "PBCI" "10jan2006" 252 "09jan2006" "13jan2006"
1424 "PBCI" "13jan2006" 223 "09jan2006" "13jan2006"
1424 "PBCI" "13jan2006" 173 "09jan2006" "13jan2006"
1424 "PBCI" "13jan2006" 156 "09jan2006" "13jan2006"
1428 "PBCI" "10jan2006" 208 "11jan2006" "11jan2006"
1428 "PBCI" "10jan2006" 226 "11jan2006" "11jan2006"
1428 "PBCI" "10jan2006" 242 "11jan2006" "11jan2006"
1428 "PBCI" "10jan2006" 279 "11jan2006" "11jan2006"
1428 "PBCI" "11jan2006" 222 "11jan2006" "11jan2006"
1428 "PBCI" "11jan2006" 184 "11jan2006" "11jan2006"
1428 "PBCI" "11jan2006" 204 "11jan2006" "11jan2006"
end
bys pid date: gen id = _n
reshape wide bg , i(pid date ) j( id)
l, sepby(pid)
On Tue, Mar 3, 2009 at 6:05 PM, Alan Neustadtl <[email protected]>
> I am using Stata V10 and have a dataset with repeated blood glucose
> measurements per day. The data are ordered by patient id (pid) and
> within patients by the date of the glucose readings (datestamp). In
> other words, there is one record per patient/blood glucose reading.
>
> I would like a new dataset with one record per patient id/date but
> with one variable for each reading that day. For example, in a month
> with thirty days where a patient took at least one blood glucose
> measurement per day, there would be thirty records. There would be a
> variable corresponding to each blood glucose measurement.
>
> The following data indicate what I currently have and what I would
> like to have. I have put a blank line between groups to make the data
> more readable but these breaks do not exist in the Stata dataset.
>
> This strikes me as a -reshape- problem but the dates are duplicates so
> -reshape- fails. If this is not a reshape problem I don't even know
> what to call it to search the Statalist archives.
>
> Any suggestions/pointers are welcome.
>
>
> ORIGINAL DATA
> pid site datestamp bglevel enrolldate fuploaddate
> 1424 PBCI 10jan2006 253 09jan2006 13jan2006
> 1424 PBCI 10jan2006 185 09jan2006 13jan2006
> 1424 PBCI 10jan2006 252 09jan2006 13jan2006
>
> 1424 PBCI 13jan2006 223 09jan2006 13jan2006
> 1424 PBCI 13jan2006 173 09jan2006 13jan2006
> 1424 PBCI 13jan2006 156 09jan2006 13jan2006
>
> 1428 PBCI 10jan2006 208 11jan2006 11jan2006
> 1428 PBCI 10jan2006 226 11jan2006 11jan2006
> 1428 PBCI 10jan2006 242 11jan2006 11jan2006
> 1428 PBCI 10jan2006 279 11jan2006 11jan2006
>
> 1428 PBCI 11jan2006 222 11jan2006 11jan2006
> 1428 PBCI 11jan2006 184 11jan2006 11jan2006
> 1428 PBCI 11jan2006 204 11jan2006 11jan2006
>
>
> NEWLY SHAPED DATA
> pid site datestamp enrolldate fuploaddate bg1 bg2 bg3 bg4
> 1424 PBCI 10jan2006 09jan2006 13jan2006 253 185 252 .
> 1424 PBCI 13jan2006 09jan2006 13jan2006 223 173 156 .
>
> 1428 PBCI 10jan2006 11jan2006 11jan2006 208 226 242 279
> 1428 PBCI 11jan2006 11jan2006 11jan2006 222 184 204 .
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
