| Variables missed after appending two de-normalized DHS datasets [message #9632] |
Tue, 26 April 2016 15:32  |
AmsP
Messages: 24
Registered: April 2016
|
Member |
|
|
Hello,
I need to de-normalize and then append two DHS datasets (e.g. BOIR31 and BOIR41). I first generate a new variable dv005=v005*(number of women aged 15-49 of the country in the survey year)/(number of women in this survey) in the two datasets, respectively. Then I append BOIR41 into BOIR31 (so BOIR31 is the basis). But after appending, all values of dv005 in BOIR41 become ".", while values of dv005 in BOIR31 are correct.
I tried several other datasets, but I always faced this problem.
Thank you very much!
|
|
|
|
| Re: Variables missed after appending two de-normalized DHS datasets [message #9633 is a reply to message #9632] |
Tue, 26 April 2016 17:20  |
Reduced-For(u)m
Messages: 292
Registered: March 2013
|
Senior Member |
|
|
Is this an append or a merge? If you are appending, the only thing I can think of is that you are not creating the same variable "type" in both datasets. If you put some actual code up (not too much, just the relevant parts) I might see something, but it seems like either you are mixing two different types of variables (one maybe is a string and the other a number) or appending in some funny way.
|
|
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
