> -----Original Message-----
> From: Sophie Barthel [mailto:[email protected]]
> Sent: Wednesday, February 19, 2003 4:09 PM
> To: [email protected]
> Subject: st: joining variables
>
>
>
> Dear List
>
> I have got the following problem:
>
> From a set of observations I have created a variable to10la for people
> whose left eye was measured and a variable to10ra for people
> whose right
> eye was measured. Whenever a person appears in the data whose
> right eye
> was measured to10la displays a missing value and vice versa. All of
> these are meant to be merged into one variable, however I do
> not seem to
> be able to find out how to do this.
>
> Thank you
>
> Sophie
You have a couple of options.
Assuming that each person appears only once, and that each person has
only one rating (eihter right or left, not both), then:
. gen rating = to10ra
. replace rating = tot10la if rating>=.
Or, equivalently
. egen rating=rmax(to10ra to10la)
which gives you the maximum of the two variables (ignoring blanks). You
would get the same thing with
. egen rating=rmin(to10ra to10la)
or with
. egen rating=rmean(to10ra to10la)
If some cases have both ratings, then you could use one of the -egen-
statements, depending on how you wanted to combine the two.
If you might have multiple records for an individual (one with a left
rating, the other with a right rating), then you should -reshape- the
data first.
Nick Winter
*
* 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/
