[EpiData-list] Summing a number of variables

epidata-list at lists.umanitoba.ca epidata-list at lists.umanitoba.ca
Fri Aug 3 15:27:43 CDT 2007


The easiest way of summing:
> I would like to make an automatic sum (see: variable "daystotadm") of a
> series of variables (see: variable "daysadm1-2..."), but if only one of
> these variable is empty, the automatic calculation using the function SUM
> give a null result.
>   
What I would do is:

dateexit1
  AFTER ENTRY
    if dateexit1 = . then 
      goto daysadm1
    endif
  end
end

dateexit2
  AFTER ENTRY
    if dateexit2 = . then 
      goto daysadm
    endif
  end
end

etc for the other dateexit3 to 5.

daysadm
 before entry
  * assume you have at least data in the first var:
  daysadm=(dateexit1-dateentry1)
  if dateexit2 <> . then
    daystotadm =(dateexit2-dataentry2) + daysadm
  endif
  if dateexit3 <> . then
    daystotadm =(dateexit3-dataentry3) + daysadm
  endif
  if dateexit4 <> . then
    daystotadm =(dateexit4-dataentry4) + daysadm
  endif
  if dateexit5 <> . then
    daystotadm =(dateexit5-dataentry5) + daysadm
  endif
 goto ... next variable
 end
end


In this way when the first dateexit is equal to ., the program jumps and 
no entry is needed in ... up to number 5 variable.
(I hope the comparison of date variable with . works - did not test it).

Regarding the other question from yesterday. Installation of EpiData 
software (entry and analysis) is always very "clean", meaning not 
disturbing the PC, so you can have many copies if you wish for various 
purposes. Just copy the exe file (plus help if needed). When testing I 
tend to create the default for testing as: c:\epidataanalysistest  such 
that the test version of analysis does not interfere with the release 
version.

regards jens lauritsen
EpiData Association


More information about the EpiData-list mailing list