Figuring out sample size for the different type of study is a tough job without the help of statistician. So is there any alternative for the researcher?
You can use G*Power (http://www.gpower.hhu.de/).
It requires a registration but is is completely free for academic use.
Please be aware that prospective power analysis is not fool proof, but depends on your estimate of the variance you expect, and thus the sample size at which you estimate this prospectively. In many fields of biology at least getting a reliable estimate of the variance in the dependent variable you expect is very limited, because it depends on the situation you measure it in etcetera. Therefore prospective power analysis becomes little more that helping you educate your guess only slightly (in many cases).
Also in most cases retrospective power analysis is wrong, because the estimate of variance you use to retrospectively calculate the power you had is confounded in the same sample you want to estimate the power for.
Yes G Power is useful. R also has some packages that could help you out if you want to estimate the power from mixed models etc.
For Power Analysis you may use the "pwr" package in R, developed by Stéphane Champely
function power calculations for
---------- ------------------------------
pwr.2p.test two proportions (equal n)
pwr.2p2n.test two proportions (unequal n)
pwr.anova.test balanced one way ANOVA
pwr.chisq.test chi-square test
pwr.f2.test general linear model
pwr.p.test proportion (one sample)
pwr.r.test correlation
pwr.t.test t-tests (one sample, 2 sample, paired)
pwr.t2n.test t-test (two samples with unequal n)
more details: http://www.statmethods.net/stats/power.html
For straight forward projects maybe this site will be helpful:
http://biostat.mc.vanderbilt.edu/wiki/Main/PowerSampleSize
Not only you get an analysis but you also get a write up of the methods. But talking with a statistician might still be helpful to help you decide that the methods being used are correct for your problems.
You can see the formula for determining the sample size in this article:
Article How to Calculate Sample Size for Different Study Designs in ...
- Badges
- Science topic
- Similar topics
- Methodology
- Research Methodology
- Research Papers