Overview for Analysis of Means

Use Analysis of Means to determine whether the mean of each group differs from the overall mean. Your data can follow a normal, binomial, or Poisson distribution. Analysis of means is a graphical alternative to ANOVA. For more information about how analysis of means compares to ANOVA, go to Understanding Analysis of Means.

For example, an engineer at a manufacturer performs an experiment to assess the effect of three process time levels and three strength levels on product density. The engineer uses analysis of means to examine the main effects and interaction between the factors.

Where to find this analysis

To perform an analysis of means, choose Stat > ANOVA > Analysis of Means.

When to use an alternate analysis

If your design has two categorical factors and is unbalanced, use Fit General Linear Model if you have all fixed factors or Fit Mixed Effects Model if you have random factors.
If your design has more than two categorical factors or includes covariates, use Fit General Linear Model if you have all fixed factors or Fit Mixed Effects Model if you have random factors.