[EpiData-list] Complete repeated information

epidata-list at lists.umanitoba.ca epidata-list at lists.umanitoba.ca
Thu Nov 29 08:54:30 CST 2007


Dear colegues of epidata,

Most current data structure for survival analysis have repetitive subjects
with differenta dates. That means that the data has more than one (say five)
subject number 8 (for example), and each line  number 8 has dates showing
diffenret periods (say period 1 to 2; 2 to 3 and so on).

This structure is necessary for cox models with variables changing over time
and more complex marginal models.

However some variables remain with the same value over time, such as sex.

I would like know if it is possible to implment in epidata entry chek file
some comand that would find previous repetitive subject and type in the
field this non changing variables over time.

That is... suppose Im typing a record with id number 8 which is data from
period 4. This command would look for the privous id 8 (just as autosearch)
form period 3 or 2 or 1, if it finds any it would sugests ou type the same
sex in this record as those found in privious periods of the same id.

Is this possible?

-- 
May the force be with you
Abraço forte e que a força esteja com você,
Pedro Emmanuel


More information about the EpiData-list mailing list