If previous studies showed a reduction of 0.71% in Hba1c (SD = 0.55) in inetrvention group and 0.25% (SD=0.47) in control group what will be the required sample size for this study to achieve 0.05 level of signficance and 95% statistical power
There is no fixed sample size. It depends upon several factors like precision of the findings, expenditure involved, time available for data collection, the characteristics of the population under study, the seriousness of the findings to be obtained etc. Since the concerned population under study i a clinical trial in infinite, sample size is to be sufficiently large. The process is like this:
Select a sample and analyze the data.
Increase the sample size and analyze the data.
Compare the findings of the two analysis.
Continue the process.
When the findings of successive two analysis become identical stop increasing sample size.
Dear Ehab,
Unfortunately I don't think it is possible to answer your question based on the information you have provided. You need to calculate the effect size of the previous intervention and also provide the ratio of the experimental and control group sizes. You should then be able to calculate the required sample size, assuming that you are using an independent samples t-test. For example, you can use http://biomath.info/power/ttestnoninf.htm.
dear peter, thank you for answer
missing data: ratio is 1 : 1
how can I calculate effect size of previous intervention ? I mentioned already educational intervention in previous studies lower HbA1c by 0.71% in intervention group and 0.25 in control group
So now can you calculate sample size for me
Peter Samuels explained the answer with reasonable ground.
I do agree with him.
The simplest and most straightforward effect size measure is the difference between two means. But the limitation of this measure as an effect size is accuracy. It’s just hard to evaluate. Instead, the Standardized effect size which is obtained by dividing the difference by either the standard deviation of the control group (giving you a Glass’s Δ) or by the pooled standard deviation of both groups (giving you Cohen’s d or Hedges’ g depending on the pooling equation used) can be used. Standardized effect sizes are designed for easier evaluation. They remove the units of measurement, so you don’t have to be familiar with the scaling of the variables.
I am not sure why you’re going for 95% statistical power, mostly it is 80% and sometimes it is 90%. As it is RCT it is unethical to just select a sample and analyze the data. Do a proper calculation prior to start.
Dear Kandasamy thank you for answer, but can you calculate me the required sample size if we assume stistical power of 90%? According to my calculations it will be 62 patients (31 in each group) is that number true?
Thanks to Björn Holzhauer for recommending my answer.
Best wishes to him.
Thanks to Justine Leenders for recommending my answer. Best wishes to him.
I agree with Kandasamy Ravichandran.
I also like to say that I am not sure why you’re going for 95% statistical power, mostly it is 80% and sometimes it is 90%. As it is RCT it is unethical to just select a sample and analyze the data.
Thanks to Ehab Mudher Mikhael for acknowledging the answers.
You can benefit from the following links.
1. Davies MJ, Bergenstal R, Bode B, et al. Efficacy of Liraglutide for Weight Loss Among Patients With Type 2 DiabetesThe SCALE Diabetes Randomized Clinical Trial. JAMA. 2015;314(7):687–699. doi:10.1001/jama.2015.9676
2. Davy BM, Winett RA, Savla J, Marinik EL, Baugh ME, Flack KD, et al. (2017) Resist diabetes: A randomized clinical trial for resistance training maintenance in adults with prediabetes. PLoS ONE 12(2): e0172610. https://doi.org/10.1371/journal.pone.0172610
3. Naeimeh Atabaki-Pasdar, Mattias Ohlsson, Dmitry Shungin, Azra Kurbasic, Erik Ingelsson, Ewan R. Pearson, Ashfaq Ali & Paul W. Franks.Statistical power considerations in genotype-based recall randomized controlled trial. Scientific Reports volume6, Article number: 37307 (2016).
4. Röhrig B, du Prel J-B, Wachtlin D, Kwiecien R, Blettner M. Sample Size Calculation in Clinical Trials: Part 13 of a Series on Evaluation of Scientific Publications. Deutsches Ärzteblatt International. 2010;107(31-32):552-556. doi:10.3238/arztebl.2010.0552.
5. Leon AC. Implications of Clinical Trial Design on Sample Size Requirements. Schizophrenia Bulletin. 2008;34(4):664-669. doi:10.1093/schbul/sbn035.
6. Mesut Mutluoglu.Hyperbaric Oxygen Therapy in the Treatment of Ischemic Lower-Extremity Ulcers in Patients With Diabetes: Results of the DAMO2CLES Multicenter Randomized Clinical Trial. Diabetes Care 2018;41:112–119 Diabetes Care 2018;41:e60.
The links given by Entedhar Sarhat may be helpful.
I like to suggest the book
"Sampling Technique'
by Cochran.
This is a good book for the topic related to the problem.
Dear Adrian, I want to know any equation to calculate my sample size, if you know one share it here with its reference
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size