Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

Re: st: Merge Append which one

From   Kay Walker <>
Subject   Re: st: Merge Append which one
Date   Sat, 06 Nov 2010 16:15:36 +1030

Hi Nirina

Re: "data in fifteen separate files" etc

1. First create a single file for each country using Merge (combining
household, district and village data); then create the variable
"Country"in each of these merged data files.

2. "how do I give a prefix to all the variables for each country to say
fr, en, us (to france, england and USA)?"
  I would not bother to alter all the variables as later you would want
them to be combined again for further analysis. I would create a
classification column in ALL the single country files named "Country".
Make all the country names available as categories,so the variable is
"Country" and the choices would be "USA", "Fr", "UK" etc.

3. Finally, when you have checked that ALL the variable names in each
country are identical, you can APPEND the countries together. Then you
should be able to perform various analyses using REGRESSION etc to
discover any differences/similarities across countries at the different
levels of Village, District etc.

If you have the same Variable names in the Village, District etc data
files, you might also need to create a categorical variable such as
"Level" with categories Village, District etc, to unite them before
Merging into the single Country files. It is difficult to say without
seeing your data file layout.


*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index