Technical Details for Choice Modelling Utility Calculations

From Q
Jump to: navigation, search

Input data

This QScript uses data from individual-level parameter means from a choice modeling experiment. The variables should be combined into a Number - Multi question that has the string Individual-Level Parameter Means in the name. The individual-level parameters question can be generated in Q from a segmentation of an experiment question (by right-clicking a node of the tree and selecting Save Individual-Level Parameter Means and Standard Deviations).

Alternatively, the individual-level parameter means can be computed externally and imported into Q as part of a larger data set. In this case:

  • There should be one variable for each attribute level.
  • Each variable should have a Label in the form: <Attribute Name>: <Level Label>
  • The variables should be combined as a Number - Multi question.
  • The text Individual-Level Parameter Means should be included in the question name.

Numeric attributes

Numeric attributes, for example Price, will be treated as categorical attributes for the purpose of this calculation. When numeric attributes are present the QScript will prompt you to specify the levels that you want to display. You should enter two or more numbers separated by commas. For example, if your attribute is Price and you want to display levels corresponding to $1, $1.50, and $2, you should enter 1, 1.5, 2 into the prompt.


A new Number - Multi question is generated in the project, and a table is added to the report to show the average of the computed values.

Personal tools