| Strata variable for Sierra Leone 2013 IR dataset missing [message #24741] |
Sun, 03 July 2022 23:07 |
Jenna
Messages: 6
Registered: September 2021
|
Member |
|
|
I am using the Sierra Leone (SL) 2013 IR (women) file. Because it is a survey data, I have to use the svy command. The strata variables v022 and v023 are similar but contain numbers from 1 to 27 instead of the usual region + urban/rural combination. DHS recommends generating a strata variable using v024 (region) and v025 (type of place of residence). However, in the case of Sierra Leone, stratification was done using districts (not regions) and type of place of residence (urban/rural). The dataset does not have a district variable. can I use v022/v023 in the svy command even though it is made up of numbers only? If not how can I create a strata variable in this case? Thank you in advance for your response. In the end I need to append the SL datasets to two other datasets (Ghana and Liberia), both of which have a clear strata variable (v023=region + type of place of residence). If I consider v022 to be the strata variable of SL, I run into issues of mismatching once I append the 3 datasets (i.e. the strata variable of Ghana gets assigned to some observations of Liberia and vice versa).
[Updated on: Sun, 03 July 2022 23:18] Report message to a moderator |
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
