http://www.stat.ucla.edu/~dinov/ |
Analysis of variance (ANOVA) performs comparisons like the t-Test, but for an arbitrary number of factors. Each factor can have an arbitrary number of levels. Furthermore each factor combination can have any number of replicates. ANOVA works on a single dependent variable. The factors must be discrete. The ANOVA can be thought of in a practical sense as an extension of the t-Test to an arbitrary number of factors and levels. It can also be thought of as a linear regression model whose independent variables are restricted to a discrete set.