Menu ENN Search
Language: English Fran├žais

How to determine sample size & the type of soft ware appropriate for my study analysis.

This question was posted the Assessment and Surveillance forum area and has 2 replies.

» Post a reply

Anonymous 1727

student of MPH in reproductive health

Normal user

15 Nov 2012, 07:37

1. I am about to begin a research on IYCF practices in one district, Ethiopia. There is no similar study deployed in the area to find the population proportion for the study & that is the reason why I need your recommendation how to fix population proportion & standard deviation to calculate the sample size. 2. Could you please suggest the type of soft ware for sample size determination & analysis of such type of study?


Normal user

15 Nov 2012, 09:42

I would like to have your suggestions by level of responsibility (Ministry of Health, management of nutrition, the regional directorates, NGOs and UN agencies).
I am also interested in helpful links related to this issue.
Thank you for your response!

Mark Myatt

Frequent user

15 Nov 2012, 11:46

(1) You usually only need the expected proportion when calculating sample sizes for proportions. You may want to use results from a MICS or DHS survey to find expected proportions. Sample sizes for the commonly used "WHO" consensus indicators can be difficult as there ares many indicators with different expected proportions and different denominators. Some indicators (e.g. continuing BF) have a narrow age-range (3 months) for the denominator. This means that you may need a very large overall sample size in order to estimate some indicators with useful precision. The "WHO" indicators work well with large samples but are not well suited to small sample surveys. You may want to consider alternatives to the "WHO" indicators. An example can be found here. (2) ENA for SMART has a sample size calculator. I commonly use GNU SampSize for sample size calculations. I have a Windows program here which you may find useful. Software for survey analysis depends on your survey design and your experience. You could use (e.g.) EpiInfo, SPPS (with the "complex samples" module), R (with the "survey" library), STATA, SAS, SUDAAN, &c. I hope this helps.

If you have any problem posting a response, please contact the moderator at

Back to top

» Post a reply