In addition, -multencode- from SSC offers a way to -encode- several
string variables at once, with the key twist that the same set of value
labels is used throughout.
Nick
[email protected]
Martin Weiss
-replace-ing these variables with numeric counterparts would require you
to
change the -type- from string to numeric as well. I think the best way
to do
this for many variables is to employ a loop over them, and have the old
ones
-drop-ped automatically...
**************
clear*
input str1 alc_now
y
y
y
y
y
y
n
y
end
clonevar alc_now1=alc_now
clonevar alc_now2=alc_now
clonevar alc_now3=alc_now
qui ds, has(type string)
foreach var in `r(varlist)'{
encode `var', gen(`var'_new)
drop `var'
}
list, noobs
**************
Michael McCulloch
I have a long list of string variables which contain only "y", "n", or
are missing.
I'm converting them to new numeric variables with values 1 if the old
variable is "y", and 0 if "n".
For example:
gen etoh_now=alc_now=="y"
This gives, correctly:
.. list alc_now etoh_now, clean
alc_now etoh_now
1. y 1
2. y 1
3. y 1
4. y 1
5. y 1
6. y 1
7. n 0
8. y 1
Is there a more efficient way to do this, where I can change "y" to 1
and "n" to 0 without creating a new variable and manually dropping the
old variable?
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/