Training

When analyzing data, your goal is to understand the relationships among variables. Most of the time, you are interested in obtaining descriptive statistics, like the mean and standard deviation, as well as performing inferential statistics like t tests or ANOVA.

For example, you may want to use Clementine to assess the relationship between a person’s income and one of three variables that signifies whether they represent a good risk, a bad profit, or a bad loss. The example below uses a Means node:

• The people who represent a bad loss have a mean income of $25,440
• Those that represent a bad profit have a mean income of $22,779
• The people who represent a good risk have a mean income of $34,122

The mean incomes for the bad loss and bad profit groups seem similar; however, those that represent a good risk have a higher mean income than the other groups (the importance column confirms this statistically). As a result, you can use the variable, income, to predict those that are a good risk.

Since descriptive and inferential statistics only show us part of the relationship between the variables income and risk, the histogram below further teases apart relationships between them:

You can see that those people with higher incomes are associated with being a good risk (just as our statistics had suggested), and those people with lower incomes are associated with being a bad profit (something our statistics were unable to capture). As a result, the variable income can also help you predict those that are a bad profit (it is only the bad loss group that is not associated with income). Clementine is equipped with several graphs that are especially powerful in helping you better understand relationships among variables, so have fun.

We offer SPSS courses at locations around the world.
Find a course in the location nearest to you.