[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
RE: st: < and > operand in recode
thanks dick for your pointers about recode dangers. following [R] recode, the included example did append a generate newvar. i think, as you warned, generating a new variable is a good practice as was tabulating against original, representing safety mechanisms should the recode produced unexpected results.
i should say that i have Stata 8 while, i think, [D] was only initiated from Stata 9 (either that or I have misplaced my [D]).
> Date: Sun, 20 Aug 2006 12:10:49 -0500
> To: email@example.com
> From: firstname.lastname@example.org
> Subject: Re: st: < and > operand in recode
> I was in the midst of posting a response to this query when Nick Cox's very
> complete answer arrived. Just two additional points. The full documentation
> for recode can be found in [D}, data management.
> The key idea is that recode rules are evaluated left to right. Thus
> (1/5=1)(5/10=2) results in a value of 5 being recoded
> to 1 even though "5" appears in both rules. Nick is certainly right;
> recoding continous variables can sometimes
> result in unexpected outcomes. At the very least, it is a good idea to
> always use the -gen- option with recode rather than
> overwrite the original variable and to immediately cross tabulate the
> original variable against the new.
> Richard T. Campbell
> Division of Epidemiology and Biostatistics
> School of Public Health &
> Institute for Health Research and Policy
> University of Illinois at Chicago
> * 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/
Try Live.com - your fast, personalized homepage with all the things you care about in one place.
* For searches and help try: