Menu ENN Search
Language: English Fran├žais

Post a reply: Clarification on SMART survey cluster numbering

Use of en-net is subject to the Terms and Conditions.

We have conducted a SMART survey and I would like to seek expert opinion on a technical question regarding the analysis for our SMART survey data.

 

 

 

 

 

 

SMART uses PPS to assign groups of households to different locations, so for example you have cluster numbers 8, 9, 10…each representing 10 households from the different locations. A particularly large location may be assigned multiple cluster numbers, and therefore more households (say for example Village Shawar gets clusters 19, 20 and 21 together, and therefore 30 households total).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

The question is: Will it affect the analysis (confidence intervals mainly) if we are assigning households different cluster IDs, even if they are technically in the same geographic area/large village? As in the example of Village Shawar, should all households in Village Shawar be assigned cluster ID 19…or they retain the original numbers from sampling (19, 20, 21).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Looking forward to your feedback and clarification on this.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Thanks, 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

To add a URL link click the chain button within the editor and enter a URL
Please give your real name

Your email will not be displayed on the website
Optional

Your question will be manually approved before it appears on the website. We reserve the right to delete any inappropriate messages.