| Grouping ethnic categories for Nigeria DHS 2008 and 2013 [message #14028] |
Sun, 04 February 2018 18:55  |
catherino222
Messages: 4
Registered: February 2017
Location: London ON
|
Member |
|
|
I am working with the Nigeria DHS of 2008 and 2013, and I want to pool the two datasets. However, ethnicity was coded differently. Ethnicity for 2008 has 11 categories while 2013 has over 200. I would like to know how to group ethnicity for 2013 to produce a final list of major ethnic groups that are comparable with those for 2008.
Thank you
Catherine
|
|
|
|
| Re: Grouping ethnic categories for Nigeria DHS 2008 and 2013 [message #14043 is a reply to message #14028] |
Tue, 06 February 2018 08:33  |
Bridgette-DHS
Messages: 3208
Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS Stata Specialist, Tom Pullum:
As you say, the 2008 survey (NG53) has two variables for ethnicity. In the IR file, the variable with 11 categories (one of which is "others") is v131. The other variable, s119, has MANY categories. You can get all the category numbers and labels with "label list v131 s119" and you can find which categories of s119 were put into which categories of v131 with, for example, "tab s119 if v131==1, nolabel".
The 2013 survey (NG6H) appears only to have the detailed list, as you note, and that's given as v131. You can get all the category numbers and labels with "label list V131". It's unfortunate that there is not a grouped list. However, if you look at how the categories of s119 were mapped into v131 in the 2008 survey, you should be able to apply the same mapping or recode to v131 in the 2013 survey. I believe the codes for the detailed ethnic categories were not consistent, so be careful....
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
