Database Reference
In-Depth Information
13.
Click
One-Way ANOVA
to select it.
In the information pane, you have a fairly detailed description of how to use this test.
14.
Click
OK
to close the dialog box.
15.
In the work area, click the
One-Way ANOVA
container to select it.
16.
In the
Properties
pane, change the
Query
property to
Statistics_query
.
17.
Change the
Bonferroni
property to
Ye s
.
The Bonferroni range test compares the different campaigns in an attempt to find which
campaigns have means that are significantly different from each other.
18.
From the
Data Items
tab of the
Insertable Objects
pane, drag the following items into
the statistics container:
•
Gross profit
into
Dependent variable
•
Campaigns Only
into
Independent variable
•
Order number
into
Cases variable
19.
From the
Run
menu, select
Run Report - HTML
to compare the dashboard to what's
shown in Figure 6.49. When prompted, select members
2004
,
Camping Equipment
,
and
Asia Pacific
, along with the Analysis of Variance.
With a significance of .000, the ANOVA shows that there are significant differences
between the means of orders for all promotions. The Bonferroni post hoc range test
shows which campaigns performed significantly better than others. For example, the
TrailChef Campaign did significantly worse than the Rising Star Campaign when it
comes to generating a higher gross profit.
20.
Close
IBM Cognos Viewer
to return to your query design.
Step 11: Create the Box Plot Chart
For the next type of statistical analysis, we want to show individual orders whose gross profit sig-
nificantly exceeds others within each campaign.
1.
In the
Explorer Bar
, mouse over the
Page Explorer
tab and select the
Report Pages
folder.
2.
In the work area, copy and paste the
Analysis of Variance
page.
3.
With the
Analysis of Variance1
page selected, double-click the
Render Variable
prop-
erty in the
Properties
pane.
The Render Variable dialog box is displayed.
4.
In the
Render for
pane, remove the check for option
2
.
5.
Add a check for option
3
.
