The table of Bonferroni confidence intervals shows the following:

·    N-The number of observations for each group formed by each combination of factor levels.

·    StDev-The standard deviation for each of group.

·    Confidence intervals-Simultaneous confidence intervals (CIs) for each group. Each CI is a range of likely values for the standard deviation of the corresponding population (also called sigma or s). You do not know the true value of s, but the CI allows you to guess its value based on the sample data. The CIs are adjusted to maintain the appropriate familywise confidence level.

·    Individual confidence level-The actual confidence level for each individual confidence interval. If the individual confidence level is close to the upper limit of 100%, then the CIs can be very wide.

The Bonferroni CIs are not the same as the multiple comparison intervals that are displayed on the summary plot. Use each type of interval as follows:

·    Use the Bonferroni CIs to simultaneously estimate the standard deviation of each population.

·    Use the multiple comparison intervals to determine which standard deviations are significantly different from each other.

For the driving data, the first factor is Experience and the second is RoadType. There are four observations in each of the cells for the six factor level combinations. The first standard deviation (StDev), 5.88784, is for Experience = 0 (inexperienced driver) and RoadType = 1 (first class road).

The Bonferroni simultaneous CIs estimate the standard deviations jointly for all of the populations. The confidence level of 95% means that there is only a 5% chance (100% - 95% = 5%) that one or more of the population standard deviations do not fall within the corresponding CI.

The individual confidence level of approximately 99.17% means that for any one of the intervals, ignoring the others, there is only a 0.83% chance that the population standard deviation does not fall within the CI.