The DHS Program User Forum      
Discussions regarding The DHS Program data and results
Home » Countries » Ethiopia » About Ethiopian DHS 2011 sample design section (pp.7) and Parent raw data set inconsistence
About Ethiopian DHS 2011 sample design section (pp.7) and Parent raw data set inconsistence [message #11858] Tue, 21 February 2017 12:02 Go to next message
sko16 is currently offline  sko16
Messages: 12
Registered: June 2016
Location: Addis Ababa, Ethiopia
Dear DHS Experts
In 2011 Ethiopian DHS the final report on page 7 in sample design section stated that " The 2011 EDHS sample was selected using a stratified, two-stage cluster design, and EAs were the sampling units for the first stage sample included 624 EAs, 187 in urban areas and 437 in rural areas". My question is about the raw data set the cluster number (V001) on individual parent data set (women data set). That the total numbers are 650 in parent raw data set. In reality the following cluster numbers are missed or Clusters not selected on EDHS 2011 parent data set ( 11, 13, 15, 31, 45, 52, 70 , 102, 109, 127, 129, 134, 144, 146, 163, 168, 175, 178, 190, 208, 211, 216, 220, 221, 237, 256, 258, 263, 298, 304, 335, 338, 384, 387, 393, 410, 424, 452, 459, 486, 489, 496, 500, 507, 524, 527, 537, 540, 565, 566, 569, 571, 574) which are 54 in number. As a result the remaining number of cluster is 650 54 = 596 or 28 clusters are not included to say 624 EAs/clusters. So, would you told me why this is the reason, if I didn't make confusion on my question? Please, don't hesitate to ask the question again if you have confusion concering this.

Re: About Ethiopian DHS 2011 sample design section (pp.7) and Parent raw data set inconsistence [message #11872 is a reply to message #11858] Wed, 22 February 2017 12:36 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 1315
Registered: February 2013
Senior Member
Following is a response by Senior DHS Sampling Specialist, Ruilin Ren:

The Ethiopia DHS 2011 was originally designed for 650 clusters. After the selection of the sample clusters, the survey was redesigned for some reasons, and 26 clusters were dropped. So the Ethiopia DHS 2011 actually included 624 clusters as stated in Appendix A. However, during the survey, another 28 clusters could not be interviewed due to several reasons, so the actual and final number of clusters included in the survey was 596.

You are right in stating that you found 54 missing cluster numbers. The clusters were numbered 1 through 650 at the onset of the survey, but the cluster numbers were not re-assigned after exclusion of clusters before data collection. As a result, the cluster numbers in Ethiopia DHS 2011 were not continuous.

We may need to improve the accuracy of Appendix A, by indicating changes made, and also reporting the survey results (number of clusters not interviewed), in the write up of Appendix A.

Previous Topic: About release of 2016 Ethiopian DHS
Next Topic: About the number of clusters selected for 2011 EDHS
Goto Forum:

Current Time: Sun Apr 22 06:26:22 Eastern Daylight Time 2018