The DHS Program User Forum: Bangladesh » Calculation of sample size in BDHS 2014

Home » Countries » Bangladesh » Calculation of sample size in BDHS 2014

Show: Today's Messages :: Show Polls :: Message Navigator

Calculation of sample size in BDHS 2014 [message #13721]

Wed, 13 December 2017 00:07

mdhasan4
Messages: 4
Registered: September 2015
Location: Bangladesh

Member

Dear Experts,

I would like to know the justification of sample size calculation. In every report, we get sampling technique what they used it but didn't mention anything about why 600 EA and selected 30 household per cluster? Is there any formula or way that can help us to convinced the people.

I am cordially requesting the experts who are involved with this program, please give your value able opinion. Sometimes we fall in big trouble, we can't give satisfactory answer to our supervisor.

[Updated on: Wed, 13 December 2017 00:07]

Report message to a moderator

Re: Calculation of sample size in BDHS 2014 [message #13731 is a reply to message #13721]

Thu, 14 December 2017 08:39

Bridgette-DHS
Messages: 3216
Registered: February 2013

Senior Member

Following is a response from Senior DHS Stata Specialist, Tom Pullum:

Under the "Publications" tab at www.dhsprogram.com you can find core documentation, including a description of the sampling procedures: https://www.dhsprogram.com/pubs/pdf/DHSM4/DHS6_Sampling_Manu al_Sept2012_DHSM4.pdf .

The number of EAs in each stratum, and the number of households in the EAs, are based on a compromise between (a) the funds and staff available for data collection and (b) the desired power of the survey to detect real differences in the population. The survey design is approved by in-country experts as well as DHS staff. You, as a researcher, should not be held accountable for the survey design.

Report message to a moderator