The Games-Howell method compares the means for each pair of factor levels using a family error rate (often called familywise error rate) to control the rate of type I error. The family error rate is the probability of making one or more type I errors for the entire set of comparisons. The Games-Howell method adjusts the individual confidence level, based on the family error rate you choose.

This method is only available if you do not assume that all populations have equal variances.

The grouping information table highlights the significant and nonsignificant comparisons. The table includes the following information.

·    Method – the multiple comparison method used to construct the family of confidence intervals from which the grouping table is generated.

·    Comparison term – each grouping information table compares levels of one factor.

·    N – the sample size for each factor level.

·    Mean – fitted means sorted in descending order.

·    Grouping – contains columns of letters that group the factor levels. Levels that share a letter are not significantly different. Conversely, if they do not share a letter, the level means are significantly different.

To determine a likely range for the difference between two means, use the interval plot for differences of means.

The grouping information for the paint hardness data shows that group A contains Blends 1, 3, and 4 while group B contains Blends 1, 2, and 3. Blends 1 and 3 are in both groups. Factor levels within a group are not significantly different from each other. Because Blends 2 and 4 do not share a letter, Blend 4 has a significantly higher mean than Blend 2.