One of my medical friend planing to study the effect of exercise and control diet on diabetic control in pre-diabetic patients? The difficulty is to decide the suitable number of subjects for getting the precise results.
Can anybody suggest how to calculate the sample size in this case? either to use the mean and SD for variables under study to calculate the sample size or to consider the proportions?
kindly elaborate in simplest way.
Dear Ali,
There is a free program available, called G*Power. You can calculated a priori the sample size you need to reach a certain statistical power, etc... with this program. Please see the link below. Beside this calculation, you should also consider the sample sizes from comparable studies.
Hope this helps,
kind regards,
Erich
http://www.gpower.hhu.de
Dear Ali,
There is a free program available, called G*Power. You can calculated a priori the sample size you need to reach a certain statistical power, etc... with this program. Please see the link below. Beside this calculation, you should also consider the sample sizes from comparable studies.
Hope this helps,
kind regards,
Erich
http://www.gpower.hhu.de
Add on to Erich:
The concept of statistical power is associated with sample size, the power of the study increases with an increase in sample size. Ideally, minimum power of a study required is 80%. The power can be estimated by using data of a pilot study or preliminary data extracted from another study, the alpha error is usually set to 0.05. Simply identify the test to be conducted in G*Power and the degrees of freedom where applicable. For your case - of I understand right - it is a rmANOVA.
Hence, the sample size calculation is critical and fundamental for designing a study protocol.
Kind regards,
Ramona
I agree with Erich and Ramona,
The article below explains not just the importance of an a priori sample size calculation, but also how to perform it considering different study designs.
Beck, TW. The importance of a priori sample size estimation in strength and conditioning research. J Strength Cond Res 27(8): 2323-2337, 2013-
Kind regards,
Antonio H. Germano-Soares
Thanks all...but I need specific solution to given situation only...If anybody having idea either should I suggest to use means of diabetic and prediabetic group or based on proportions. If we used proportions which proportion should I consider to calculate sample size for aforesaid condition..please
If you want to be sure, take 3-5% patients from that population in your sample.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size