The ANOVA F-test to compare the means of k normally distributed populations is not applicable when the variances are unknown, and not known to be equal. A spacial case, k=2, is the famous Behrens-Fisher problem (Behrens, 1929; Fisher, 1935). Welch (1951) test was proposed to fill this void, a generalization to his 1947 previous paper (Welch, 1947).
This m-file works without all the data samples. But only with the size, mean and variance samples.
Syntax: function welchanova(x,alpha)
x – data nx2 matrix (Col 1 = data; Col 2 = sample code)
alpha - significance level (default=0.05)
- Summary statistics from the samples
- Decision on the null-hypothesis tested