On May 18, 2009, at 3:40 PM, Richard Goldstein wrote:
I have a data set with 3 variables: field, cat, order; there are 258
observations
I want the 258 observations in field to be 258 variables where the
name of the value in field becomes the new variable name; I want to
keep the associated values of the other two fields
For example, say that the first observation has the following values:
field cat order
namea A 1
I want a new set of variables such that I have a variable called
"namea" and I still have cat (of "A") and order (of "1") -- but
remember that I should now have 258 variables--cat and order can be
joined as the value for each variable; thus, for the example above,
the first variable (namea) would now have the value "A1"
If you're interested in an off-the-shelf routine, then Nick's -sxpose-
will do the trick:
gen value = cat + string(order)
keep field value
sxpose, clear force firstnames
-- Phil