Fwd: Re: st: Comparing Variable Name Labels Between Datasets

[Michael Norman Mitchell <Michael.Norman.Mitchell@gmail.com>]

Dear Beth

  Referring to my reply below, if you are interested in doing #1, then 
I think that the -descsave- command could be useful. It saves a dataset 
based on the -describe- command, of which you can focus on the names of 
the variables and their labels. For each dataset, you could output such 
a dataset, and then compare the datasets with the variable names and 
names of the variable labels. You can install this via -ssc install 
descsave-.

  If you are interested in #2, then I think that the -vlc- command 
could be useful. This compares the value labels across datasets. You can 
install this via -ssc install vlc-.

  It is possible that you might want to do both of these.

I hope this is helpful.

Best regards,

Michael N. Mitchell
See the Stata tidbit of the week at...
http://www.MichaelNormanMitchell.com




-------- Original Message --------
Subject: 	Re: st: Comparing Variable Name Labels Between Datasets
Date: 	Thu, 13 May 2010 19:54:52 -0700
From: 	Michael Norman Mitchell <Michael.Norman.Mitchell@gmail.com>
Reply-To: 	Michael.Norman.Mitchell@gmail.com
To: 	statalist@hsphsun2.harvard.edu



Dear Beth

  This is quite a challenging task, but I feel that you have laid it
out in a very sensible manner. I have one question regarding "step c".

  Are you interested in

1) comparing the "names of the variable labels" between the datasets.
That is, the variable "gender" is labeled with the value label "mf" in
the first dataset. The variable "gender" is also labeled with "mf" in
the second dataset. But, "mf" is labeled with "sex" in the third dataset.

  or are you interested in

2) comparing the "values and labels" for the variable labels for the
variables across datasets. That is, the variable "gender" is labeled
with "mf" in the first dataset, and the values are "0 = male" and "1 =
female" in the first dataset. But, in the second dataset, the variable
"gender" which is labeled with "mf" is labeled as "1 = male" and "2 =
female".

  Perhaps you are interested in checking both of these?

Best regards,

Michael N. Mitchell
See the Stata tidbit of the week at...
http://www.MichaelNormanMitchell.com

On 2010-05-13 7.23 PM, Beth Gifford wrote:
>  c) compare the variable name labels across years to double check that
>  the new sensibly named variable is measuring what I think that it is
>  measuring. (HELP)

*
*   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/