hi every one,
one of the reviewer comments on my article ( my article is about genetic mutation screening) is " How was the sample size determined? What is the power of the study " please help me for answer this question.
Morning,
1. Sample h report the n in your study and the population. Then explain, how you come from population to the n-sample. You dplain, why you think your n-sample is representative for the population. There are different sampling techniques. It is a very important comment since it shows generalizability of your results.
2. Provide model fit statistics for your model as well as R-squared (adjusted) for your DVs. If your overall model fit is good (CFI, GFI, AGFI, RMSEA, Chi-Squared/df, etc, depends on your model) and if you explain enough variance in yours DVs, you research is very valuable.
Best,
Eugene
With a specified level of accuracy the minimum sample size may be specified. For instance the 95%confidence interval of mu is xbar+/_1.96*sigma /√n. Therefore with error of e, minimum n=[(1.96*sigma)/e]**2
The power of a statistical test is defined by 1-beta where beta is the probability of type 2 error. That is, beta is the probability of accepting a false null hypothesis. Therefore the power is the probability of rejecting a false null hypothesis. Power = P(accepting Ha/ Ha). This is easy to calculate under the Gaussian sampling distribution.
You can also define your sample size using GPower - a free app which calculates number of relationships and type of analysis.
http://www.gpower.hhu.de
You can find also necessary references for Legitimation on the botton of the website.
Best,
Eugene
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size