| Cannot replicate NAR for multiple Cambodia DHS surveys [message #25594] |
Tue, 15 November 2022 10:00  |
MPerrot
Messages: 1
Registered: November 2022
|
Member |
|
|
Hello,
I am working on DHS data for Cambodia, using 4 databases - the ones compiled in 2000, 2005, 2010 and 2014.
I have the following issue: when computing the Net Attendance Ratio using the DHS methodology specified on Github ( https://github.com/DHSProgram/DHS-Indicators-R/commit/90f08e 6c908fd13fee8f78939a4a0ef45454002f#diff-444cb1ceb74b4400ffeb c578d6b13b6fb4a9966d5519a6faabd324395fe95666 --> R version), I do not find results similar to those of the official final report.
I don't believe it is because of my code, I have checked for errors multiple times and also tried using the official code from Github with no additional modification than those absolutely necessary (path name...): I still do not find the correct ratios. However, by tweaking the start of school year and start of school year month imputations, I manage to find adequate results, using start of the year 2000 and start of the month 6 or 7 (for June or July). I had previously used year 1999 and month 10, as UNESCO indicates that "For primary to post-secondary education, the academic year begins in October and ends in July" (https://uis.unesco.org/en/country/kh under "Education and Literacy" > "Education system")
Therefore my question is the following: would it be possible to obtain the values used to compile DHS report for
- beginning of the school year,
- beginning month of the school year,
for all my needed Cambodia data (2000, 2005, 2010, 2014) ?
Thank you very much for your help.
M. Perrot
[Updated on: Tue, 15 November 2022 10:01] Report message to a moderator |
|
|
|
| Re: Cannot replicate NAR for multiple Cambodia DHS surveys [message #25652 is a reply to message #25594] |
Tue, 22 November 2022 16:21  |
Janet-DHS
Messages: 888
Registered: April 2022
|
Senior Member |
|
|
Following is a response from DHS Senior Technical Director, Trevor Croft:
For the 2014-15 survey, the start of the school year was year=2014 and month=10. For the 2010 and 2005 surveys, there was no adjustment for the start of the school year. Age of the household member was used directly from the household schedule. We don't have the code for the 2000 survey, but believe it was the same as for 2005 and 2010.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
