How to Create a Utilities Plot

Utilities plots are the oldest data visualization developed for conjoint analysis. They first appeared in the work of early leading light Paul Green, who referred to them as plots of part-worth.

They plot the mean utilities of variables in a Choice Model (the mean over all respondents). The mean utilities may be scaled (see scaling options below). Note that the scaled values shown would be different from the means from the utilities produced by scripts such as Choice Modeling - Save_Variable(s) - Utilities (Min 0) as the former are scaled mean utilities whereas the latter are mean scaled utilities.

The plot has a number of nice features. First, it allows us to easily see the relative importance of the different attributes, which is defined as the difference between the utility of the highest and lowest levels of each attribute. Second, it makes it easy to see the relative appeal of different attribute levels. For more reading on them see our blog post here. This article describes how to go from a choice model output created in Displayr:

To a utilities plot based on the choice model attribute utilities:

Requirements

A document containing one or more choice model outputs created in Displayr.

Please note these steps require a Displayr license.

Method

1. From the toolbar menu, select Anything > Advanced Analysis > Choice Modeling > Utilities Plot.

2. From the object inspector on the right, select the choice model output that you want to use as Input to the utilities plot. This can be one of

A Choice model such as a Latent Class Analysis or Hierarchical Bayes choice-based model.
Multiple numeric variables that each contain respondent utilities for an attribute level. For example variables created using Choice Modeling - Save Variable(s) - Utilities (Min 0).
A vector or list containing summary values for each attribute level. e.g c(`Feed: Grain` = 0.2, `Feed: Vegetables` = 0.1, `Weight: 55g` = 0.4, `Weight: 60g` = 0.5, `Weight: 75g` = 0.6) or list(Feed = c(Grain = 0.2, Vegetables = 0.1), Weight = c(`55g` = 0.4, `60g` = 0.5, `75g` = 0.6))

3. [Optional]: Select a filter or weight variable from the Filter(s) and Weight drop-down box to filter the utilities plot output.

5. [Optional]: Enter a comma-separated list of Attributes to exclude from the plot, if any.

6. [Optional]: Select the desired option from the Scaling drop-down box. Default setting is "As is".

7. [Optional]: Select the Attributes order (As is, Increasing or Decreasing).

8. [Optional]: Select the Levels order (As is, Reverse, Increasing, Decreasing).

9. [Optional]: Provide a comma-separated list of any Attributes to exclude from ordering from ordering.

10. [Optional]: Set Zero line width of the axis line at Utility = 0. Often you want to emphasize this line but it can also hidden by setting the width to zero.

Additional Properties

When using this feature you can obtain additional information that is stored by the R code which produces the output.

To do so, select Create > R Output.
In the R CODE, paste: item = YourReferenceName
Replace YourReferenceName with the reference name of your item. Find this in the Report tree or by selecting the item and then going to General > General > Name from the object inspector on the right.
Below the first line of code, you can paste in snippets from below or type in str(item) to see a list of available information.