CART® Regression for Predict Rating Scale Survey Responses

Use CART® Regression to use complex relationships with multiple predictors to predict survey responses.

This example applies to the Customer Contact Center Module. For more information, go to www.minitab.com/customer-contact-center-module.

Example

A rating scale survey question is one where the response can take on many values, such as 1–10 or 1–100.

In this worksheet, Rating is the response. Hold Time and Speed of Answer are the continuous variables. Call Center Location and Issue are the categorical variables. The predictors may explain differences in survey responses.

C1	C2	C3	C4-T	C5-T
`Rating`	`Hold Time`	`Speed of Answer`	`Call Center Location`	`Issue`
8	50	18	East	Late payment
6	45	22	East	Technical question
7	58	28	West	Incorrect charge
4	102	38	West	Technical question

How-to

Choose Solutions Modules > Functions > Customer Contact Center KPIs, then select Launch.
Under Customer Satisfaction, select Rating scale survey.
Select Predict rating scale survey responses, then click OK.
Select CART® Regression, then click OK.
In Responses, enter the column that contains the survey ratings. The response is also called the Y variable.
In Continuous predictors, enter the columns of numeric data that may explain or predict changes in survey ratings. The predictors are also called X variables.
In Categorical predictors, enter the categorical classifications or group assignments that may explain or predict changes in survey ratings. The predictors are also called X variables.
Click OK.

Tip

For more information about this analysis, click Help in the main dialog box.