Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: st: "Label define" syntax for "all other values"


From   Nick Cox <n.j.cox@durham.ac.uk>
To   "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu>
Subject   RE: st: "Label define" syntax for "all other values"
Date   Fri, 24 Sep 2010 13:26:19 +0100

Indeed. There are many possibilities here, but all I think with disadvantages as well as advantages. For example, if what is driving this is a desire for graphs and/or tables of simplified classifications, then you could write wrapper programs that produce versions of the variables in question on the fly. 

Nick 
n.j.cox@durham.ac.uk 

Dan Waldo

Thank you, Nick. Having 2 distinct approaches is a consolation for not having the 3rd approach. Creating a second variable is a clean and viable option, although I suspect its viability fades as the number of variables needing this treatment increases.

--- On Fri, 9/24/10, Nick Cox <n.j.cox@durham.ac.uk> wrote:

> The straight answer is that -label
> define- does not offer the syntax imagined. In addition to
> Eric's strategy, another is 
> 
> clonevar rr_job2 = rr_job 
> replace rr_job2 = 5 if !inlist(rr_job, 1, 2, 3, 4) 
> label def rr_job2 1 "Driver" 2 "Conductor" 3 "Brakeman" 4
> "Fireman" 5 "Stata user"
> label val rr_job2 rr_job2 
> 
> -recode- is here naturally an alternative to -replace-. 

 Dan Waldo
 
> Thanks, Eric ... Your advice has put me on the right track
> (sorry about the pun).
> 
> I see that rearranging the code one can accommodate values
> of interest that appear less neatly among the possible
> values:
> 
> ********
> levelsof rr_job , local(lev)
> foreach x in `lev' {
>     lab def rr_jobf `x' "All Other Jobs", modify
>     }
> 
> lab def rr_jobf 1 "Driver" 5 "Conductor" 6 "Brakeman" 12
> "Fireman", modify
> 
> lab li
> lab val rr_job  rr_jobf
> ********
> 
> I had thought that perhaps there was a compact way to do
> this, akin to the SAS FORMAT procedure syntax:
> value rr_jobf 1="Driver" 5="Conductor" 6="Brakeman"
> 12="Fireman" other="All other jobs";
> 
> ... but such is life.
 
Fri, 9/24/10, Eric Booth <ebooth@ppri.tamu.edu>
 
> > lab def rr_jobf 1 "Driver" 2 "Conductor" 3 "Brakeman"
> 4
> > "Fireman", modify
> > 
> > levelsof rr_job if rr_job>4, local(lev)
> > foreach x in `lev' {
> >     lab def
> > rr_jobf   `x'  "All Other Jobs", modify
> >     }
> > 
> > lab li
> > lab val rr_job  rr_jobf
 
Dan Waldo
 
> > > Suppose that variable rr_job takes the values
> > 1,2,3,...,999. Can I construct a format for this
> variable
> > that specifies some values and lumps the rest into a
> > catchall category? That is, can I write
> > > 
> > > label define rr_jobf 1 "Driver" 2 "Conductor" 3
> > "Brakeman" 4 "Fireman" <other> "All other jobs"
> > > 
> > > where <other> represents the syntax I don't
> know
> > but means any value not elsewhere specified.

*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index