I find the matter more open than is implied by this unduly stern homily.
First off, people analysing data are often not the people who produced
the data files, so you may well be haranguing the wrong person.
More fundamentally, there is a very good case for self-documenting data
files. Anyone who has been handed "the data" as one or more files that
have long since been detached from what we now often call metadata
should appreciate the point, sometimes from repeated and bitter
experience. Stata input commands have long been supportive of headers.
In this case, Joseph has yet to report on a solution and, until that is
so, we can not be clear on what really is the problem. It may turn out
to be something quite different from the existence of a header.
Nick
[email protected]
Sergiy Radyakin
and as a lesson, next time you are doing an experiment, don't place
any "header" information in the datafile.
If you need additional information related to the setup of the
experiment, to the subject, etc, have it in a separate file, and carry
an experimentID in your data files. Proper data organization from the
start will keep you from frustration later on.
*
* 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/
