Edson wrote:
When using the command "append" to combine datasets what
to do if the same variables have different alternatives
in the different survey editions (as in the example below
- 2006=4 answers and 2002=3 answers)? Is it a problem or
can be ignored?
use "C:\Documents and Settings\aksd\stata1\dataset2006.dta", clear
ta literacy
for those with codes |
10-17 in column 4 aged |
10years and abov | Freq. Percent Cum.
-------------------------+-----------------------------------
unable to read and write | 6,177 34.56 34.56
able to read only | 896 5.01 39.57
able to read and write | 10,797 60.40 99.97
uses braille | 5 0.03 100.00
-------------------------+-----------------------------------
Total | 17,875 100.00
. use "C:\Documents and Settings\aksd\stata1\2002\dataset2002.dta", clear
. ta literacy
liter | Freq. Percent Cum.
------------+-----------------------------------
1 | 7,670 30.64 30.64
2 | 1,450 5.79 36.43
3 | 15,912 63.57 100.00
------------+-----------------------------------
Total | 25,032 100.00
========================================================================
In dataset2000.dta, the variable -literacy- apparently has no value
labels. Hopefully you have access to some documentation so that
you are not forced to guess the meaning of the codes.
There is a chance (but we don't know) that -literacy- in dataset2006.dta
is coded like this:
1 unable to read and write
2 able to read only
3 able to read and write
4 uses braille
and that the codes in dataset2002.dta mean:
1 unable to read and write
2 able to read only
3 able to read and write
If you can confirm this, it is your decision what to do with the five
observations of Braille readers. You could leave them as is.
If - apart from the Braille readers - the codes in the two datasets
mean different things, you need to "harmonize" them, i.e., modify
them so that you obtain the same coding - before appending.
Useful commands for that work:
recode
label variable
label define and label values
numlabel
Hope this helps
Svend
________________________________________________________
Svend Juul
Institut for Folkesundhed, Afdeling for Epidemiologi
(Institute of Public Health, Department of Epidemiology)
Vennelyst Boulevard 6
DK-8000 Aarhus C, Denmark
Phone, work: +45 8942 6090
Phone, home: +45 8693 7796
Fax: +45 8613 1580
E-mail: [email protected]
_________________________________________________________
*
* 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/