416 Statistical Methods
3 Click the Categories button, select Cola_Type from the range names
list, and click OK.
4 Click the Output button, and send the output to a new chart sheet
named Boxplots. Click OK.
5 Click OK to create the boxplots.
6 You improve the chart by editing the labels at the bottom of the
boxplot, removing the text string Cola_Type= from each label, and
increasing the font size. See Figure 10-17.Figure 10-17
Boxplots offoam versus
cola type
From the boxplots, you can see that there are no extreme outliers evident
in the data, but there are several moderate outliers; perhaps most notewor-
thy are the outliers for regular Pepsi and regular Shasta. An advantage of the
boxplot over the multiple histograms is that it is easier to view the relative
change in foam volume from diet to regular for each brand of cola. The fi rst
two boxplots represent the range of foam values for regular and diet Coke,
respectively, after which come the Pepsi values, Shasta values, and, fi nally,
the generic values. Notice in the plot that the same pattern occurs for both
the diet and the regular colas. Coke is the highest, Pepsi is the lowest, and
Shasta and generic are in the middle. The difference in the foam between
the diet and the regular sodas does not depend much on the cola brand.
This suggests that there is no interaction between the cola effect and the
type effect.
On the basis of this plot, can you draw preliminary conclusions regard-
ing the effect of type (diet or regular) on foam volume? Does it appear that