You can use the sample size formula for simple random samples. This requires to establish the prevalence of the study variable, the confidence level of the test (95%) and the margin of error (between 3 and 7%). if you are going to compare groups, it is also necesary to consider the design effect.
Is there any link from where i can read about it....and where it is given in simple language ?
You can use this simple free program with which I have good experience.
http://www.gpower.hhu.de/
edx.org has a free course currently on Fundamentals of Clinical Trials. Week 6 is about sample size calculations. The course is finished so you will need to register soon to have access.
https://www.edx.org/course/harvardx/harvardx-hsph-hms214x-fundamentals-941
JAMA has a new series on statistics, application and interpretation, including this paper that might be of interest:
Stokes L. Sample Size Calculation for a Hypothesis Test. JAMA. 2014;312(2):180-181. doi:10.1001/jama.2014.8295.
A priori information about the prevalence of events in the control group which represents the general population is required, and the effect size of the intervention i.e. extent of reduction of the event compared to the control population has to be decided considering the clinical significance. At a 5% level of significance and 80% power, the sample size formula for comparing the difference between two proportions can be used. If a cluster sampling design is being used, we have to consider the design effect ( 2 or 1.5) in the calculation of sample size.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size