Enter a value for either the Sample Size OR the Power fields. The remaining empty field will be calculated. You can perform multiple power/sample size calculations.

Sample size is the number of observations in a sample.

Power is the probability of accepting the alternative hyptothesis when it is in fact true.

Effect Size is the measure of strength of a phenomenon (effect). See the definition box on the right-hand-size.

Significance Level is the α value

two-sample: to compare the mean value between two samples

one-sample: to compare the mean value between one sample versus a given value

paired: when observations to are performed on the samples or subjects (e.g. before and after), that is, when a one-to-one relationship exists between values in the two data sets.

two-sided: to test whether a sample is either greater than or less than a certain range of values. The hypothesis doesn’t have directionality

less or greater: (a.k.a one-sided); an inexact hypothesis in which the value of a parameter is specified as being either above or equal to a certain value, or below or equal to a certain value. The hypothesis has directionality.

The highlighted field denotes which variable has been calculated.

Effect size is assess as:

d = \frac{|\mu_{1}-\mu_{2}|}{\sigma}

Cohen suggests that d values of 0.2, 0.5 and 0.8 represent small, medium and large effect sizes respectively.

Not sure if you are running the right test? Try the Statistical Decision Tree Wizard