AnalyticBridge

Data Intelligence, Business Analytics

Measures of Association: Nominal/Continuous Variables

I have a set of roughly 20-30 nominal/categorical variables and would like to determine if a relationship exists between them and a continuous-scale variable?

Besides doing an ANOVA to test for differences in the levels of the variables, what are some other techniques you have found useful in doing this?

Thanks for any input

▶ Reply to This

Replies to This Discussion

Permalink Reply by Georgette Asherman on September 19, 2011 at 1:33pm

Anova makes the assumption that each level of a categorical variable has a normal distribution for the response variable with equal variances. Why notoriously robust, I would look at the data on a trellis-style graph before doing an ANOVA. The y-axis for each subplot will the response variable. Since it is a trellis the scale will be the same for each nominal variable. The x-axis for each subplot will be each level of the nominal variable. Nearly all popular statistics and business intelligence software provide for these graphs now.

This graph will have several uses. 1) You will see which levels of the categories are used. 2) You will see the distribution within each nominal variable. 3) You can look across the nominal variables. Then you can think about what type of modeling approaches to use.

Georgette Asherman

Direct Effects, LLC

201 673-4301

▶ Reply

RSS

Welcome to
AnalyticBridge

Sign Up
or Sign In

Or sign in with:

Top Content

1	How to develop churn prediction model for telecom company?
2	Map of Data Science University Programs
3	Use PRESS, not R squared to judge predictive power of regression
4	Text mining: alternative to Wordflex?
5	Audience expansion
6	Simple solutions to make videos with R
7	Strategy for building a “good” predictive model
8	The End of Theory: The Data Deluge Makes the Scientific Method Obsolete \| Wired
9	Is Open Source Really Free?
10	Understanding the Kalman Filter Application in Economic Time Series Data