Chapter 8 Regression and Correlation 343
tests, the Bonferroni approach rolls all of the tests into one big package, with
0.05 referring to the whole package.
Bonferroni makes it much harder to achieve signifi cance, and many re-
searchers are reluctant to use it because it is so conservative. In any case,
this is a controversial area, and professional statisticians argue about it.
EXCEL TIPS
To create a correlation matrix with Excel, you can click the
Data Analysis button from the Analysis group on the Data tab
and then click Correlation from the Data Analysis dialog box.
Complete the Correlation dialog box to create the matrix.
Creating a Scatter Plot Matrix
The Pearson correlation measures the extent of the linear relationship
between two variables. To see whether the relationship between the variables
is really linear, you should create a scatter plot of the two variables. In this
case, that would mean creating 15 different scatter plots, a time-consuming
task! To speed up the process, you can create a scatter plot matrix. In a scatter
plot matrix, or SPLOM, you can create a matrix containing the scatter plots
between the variables. By viewing the matrix, you can tell at a glance the na-
ture of the relationships between the variables.
To create a scatter plot matrix:
1 Click Multi-variable Charts from the StatPlus menu and then click
Scatter plot Matrix.
2 Click the Data Values button and select the range names ACT Math,
Alg Place, Alg2_Grade, Calc, and HS Rank.
3 Click the Output button, and send the output to the worksheet
SPLOM. Click the OK button twice.
Excel generates the scatter plot matrix shown in Figure 8-25.
•
