I shall be grateful if any expert would share us the formula to estimate the study sample for a comparative study please.
Rabiul, use G*Power, it's free:
http://www.gpower.hhu.de/en.html
Here are some examples:
http://www.ats.ucla.edu/stat/gpower/indepsamps.htm
Regards, Karin
G Power highly recommended, used myself couple of times.
Regards, Witold
I would recommend the article by :Bartlett,J.E.,Kotrlic,J.W.
, and Higgins,C.C.(2001) in Information Technology, Learning, and Performance Journal,Vol.19,N. 1
It depends on what you want to estimate. I would recommend Scheaffer, Mendenhall, and Ott, Elementary Survey Sampling, Duxbury. What you are looking for is probably in Chapter 4. Other books that cover this topic include Snedecor and Cochran, Statistical Methods, 8th ed, Iowa State and Ott and Longnecker An Introduction to Statistical Methods and Data Analysis, 6th ed, Brooks/Cole. Look in the index for sample size. Best wishes. P.S. If you choose to use a program from a website make sure it is for what you want to estimate. Many are for a SINGLE estimator, often the sample proportion.
Thank you very much to all who responded with lot of valuable suggestions and resources. But, I need a simple formula for a comparative study. I have to present it to get an approval of a study protocol that I am going to conduct soon.
Then you have to tell us what you want to estimate eg means, proportions, differences of means, etc. Once we know that and how you will collect data, we can answer the question.
Dear Dr.Rabiul Alam.After reviewing the valuable answers of the researchs who answered your question and reviewing the article I cited for you, I am confident that the article will answer your question.
Dear Rabiul Alam,
You can try prevalence based formula in which the total number (n)can be divided into control and intervention groups.
Prevalence based formula as explained by Naing, L (Naing et al., 2006) is as follows:
n =Z2 P (1 − P)/ d2
Where,
n = sample size,
Z= statistic for level of confidence (1.96 for 95%)
P= Prevalence (in proportion of one)
d= Precision (in proportion of one)
I would like to recommend use of the formula for Randomized controlled trials since you want to compare the effectiveness of two drugs
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size