How to Perform Mixed-Mode Cluster Analysis in Displayr

This article explains how to do a Mixed-Mode Cluster Analysis in Displayr. Mixed-Mode Cluster Analysis is a clustering technique that allows all available question types in Displayr as inputs to the analysis.

Requirements

A series of questions in your data set that you want to use as inputs to the Mixed-Mode Cluster Analysis. You can use any Displayr variable structure as an input to the analysis.

Method

In the Report tree, select +> Advanced Analysis > Cluster > Mixed-Mode Cluster Analysis.
OPTIONAL: If you have multiple data sets in your document, you'll be prompted to select the data set that you want to use for the analysis. If so, select a data set.
Select variables/variable sets from the Available data list on the left to include as inputs to the analysis. Selected variables or variable sets will appear on the right in the Data to display list.
Select the number of groups/clusters:
1. Work out the number of groups automatically - Displayr will automatically work out the number of clusters/segments using the Bayesian information criterion.
2. Specify the number of groups - enter the number of groups/clusters you want to create.
OPTIONAL: Select a Filter if you want to run the analysis based on a filtered subgroup of respondents.
OPTIONAL: Select a Weight to weight the input data.
Click Create Mixed-Mode Cluster Analysis.

A new single-response variable is added to the bottom of the data set, called "Latent Class Analysis," with a date/time stamp in the variable label. You can use this variable just like any other categorical variable in your analyses.

To modify the analysis, select the Mixed-Mode Cluster Analysis output and go to Properties > Data > Mixed-Mode Cluster Analysis.