G*Power 3 is a major extension of, and improvement over, G*Power 2 . It covers statistical power analyses for many different statistical tests like:

F test,
t test,
2-test and 
z test families
some exact tests.

G*Power is offering five different types of statistical power analysis:

A priori (sample size N is computed as a function of power level 1-�, significance level �, and the to-be-detected population effect size)
Compromise (both � and 1-� are computed as functions of effect size, N, and an error probability ratio q = �/�)
Criterion (� and the associated decision criterion are computed as a function of 1-�, the effect size, and N)
Post-hoc (1-� is computed as a function of �, the population effect size, and N)
Sensitivity (population effect size is computed as a function of �, 1-�, and N)

G*Power 3 provides improved effect size calculators and graphic options. It supports both a distribution-based and a design-based input mode.

What's New in This Release: [ read full changelog ]

· Fixed a problem in the X-Y plot for a range of values for Generic F tests. The degrees of freedom were not properly set in the graph, leading to erroneous plot values.