Okay, so I am doing a study on Iron Deficiency Anemia. In the first phase, I will determine the prevalence and then I need to go for Case control and an RCT. I need to ask if I have to calculate sample size for Case Control and RCT or can I have some standard minimum number of my own choice?
I mean, I have seen a lot of papers where the authors have just taken a certain number of cases and two or three controls per case. So can I take (for example) 50 cases and 100 controls, depending on my prevalence results or is there a specific formula for calculation?
Similar is the case with my Randomized Control Design, that is, can I take a certain number of people to start with or (again) is there any formula?
Thanks.
RCT and case control studies are analytic, measuring causal relationships with some studied risk factor(s). The prevalence (descriptive) does not determine the power of any such studies. The sample size required will depend on the alpha level of significance chosen , the power required and the magnitude of the likely effect measure,
there are many RG answers with directions to methods and actual calculators.
One of the most thorough expositions is Breslow and Day IARC 1987 vol 2.
Certainly true as Dr James Leigh has mentioned. Many formulas exist for Calculation of sample size depending upon what is most suitable for our study design. Reference given above gives a vast idea about statistics.
But the initial question; calculation of sample size for prevalance study, may be we can go with common formula n= (Z2 x p x q)/ d2
where Z=1.96, p= assumed prevalance (50%), q= 100-p (50%) , d= precision value (10% of prevalence i.e. 5)
roughly n= (1.962 x 50 x 50)/52
n= roughly comes to 400
Obviously the the need of statistician always exist to verify!!!!
Thanks for your help.
Yes, this prevalence formula is well known.
However, I am specifically looking for sample size for Case Control and RCT studies.
All the hypothesis testing studies (here RCT and case control studies) needs sample size estimation before start of study. Estimation procedure needs certain assumptions. For example (for case control study) you should have the desired level of power of the study (typically 0.84 for 80% power); statistical significance (typically 1.96 for 5% significance); ratio of controls to cases (typically 1: 1); percent of case/control exposed and an expected Odds Ratio. You can use an online calculator that calculates sample size.
http://www.openepi.com/SampleSize/SSCC.htm
As you can see that the Ratio of Controls to Cases is one of the factor deciding total sample size. So you can't take Arbitrarily 50 cases and 100 controls or vice versa,
Thanks so much for this great reply.
I easily calculated the sample size using this.
Thanks a lot indeed. :)
Best thing is to keep the number of cases and control equal it is not compulsory to calculate the sample size through any formula as all are adjustable (prevalence, population, error chance) and subject to bias.Main thing is that the number should be sufficient to draw a conclusive opinion. Previous similar studies can be quoted for sample size, Arbitrary number can also be taken as sample size , formula can also be used.
It is better not to mislead the people. There is contradiction in Ashique Ali statement: "....it is not compulsory to calculate the sample size through any formula as all are adjustable ...Main thing is that the number should be sufficient to draw a conclusive opinion...". I don't understand the magic behind in finding a sufficient number without using any formula.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size