How to Transform Alida MaxDiff Data for Use in Displayr

This article explains how to automatically transform a standard Alida MaxDiff data export into a structure that can be analyzed in Displayr using Hierarchical Bayes or Latent Class Analysis. 

Alida MaxDiff data is exported with one row per task and one variable for each alternative shown in that task:

The process outlined in this article transforms the Alida data by creating:

• A MaxDiff design matrix
  
   
• A respondent-level dataset with Most/least preferred variables for each task
  

This is the structure needed to run a MaxDiff analysis in Displayr. 

Requirements

• A standard Alida MaxDiff data export.

Method

Step 1 - Import the Alida MaxDiff dataset into Displayr

Import your Alida MaxDiff data export into Displayr as you would any other dataset.

Step 2 - Create a Version variable

Next, we'll create a Version variable that assigns an incrementing version number to each respondent. This will be used to create the design and will also be included in the respondent-level dataset. We do this using a JavaScript variable:

1. Under Data Sources, hover over a variable and select + > Custom Code > JavaScript > Numeric.
2. In Properties , go to Data > JavaScript Code and tick the Access all data rows option. 
3. Go to the General tab of Properties and change the Label and Name to Version.

4. Paste the code below into the Code panel.

   var result = new Array();
   var version = 1;
   var lastID = MemberID[0];
   for (var i = 0; i < N; i++) {
      if (MemberID[i] != lastID) {
          version++;
          lastID = MemberID[i];
      }
      result[i] = version;
   }
   result

5. Click Calculate.

Step 3 - Create the MaxDiff design

This step generates the MaxDiff design matrix from the Alida data:

1. Download the design template.
2. Upload the design template to your Displayr Cloud Drive.
3. In the Report tree, select + > Template and select the 'Alida MaxDiff Design' template from your Cloud Drive. This will create an output in your Report tree called md.design.widget.
4. In the md.design.widget Properties , select the following inputs:
   • Version - select the Version variable created in Step 2. 
   • Variables - select each of the alternative variables from the Alida MaxDiff dataset. 
   • Not shown label - "Did not see" is the standard label in Alida MaxDiff of the alternatives that were not shown to respondents in a given task. You can verify this by selecting any of the alternative variables and, in Properties , click Values and Labels to see each of the category labels. 
5. Click Calculate to generate the design output.

Step 4 - Create the most/least preferred dataset

This step generates the respondent-level most and least preferred variables for each task.

1. Download the data transformation template. 
2. Upload the data transformation template to your Displayr Cloud Drive. 
3. In the Report tree, select + > Template and select the 'Alida MaxDiff Data Transform' template from your Cloud Drive. This will create an output in your Report tree called md.data.widget.
4. In the md.data.widget Properties , select the following inputs:
   • MemberID - the respondent identifier variable. 
   • Version - the Version variable created in Step 2 above. 
   • Variables - the alternative variables from your Alida dataset. 
   • Most attribute label - the label used in Alida for the most preferred attribute.
   • Least attribute label - the label used in Alida for the least preferred attribute. 
   • Output file name - the name for the output file (this will be saved to your Displayr Cloud Drive) automatically. 
5. Click Calculate to generate and save the most/least dataset to your Cloud Drive.

Step 5 - Import the respondent dataset

Import the most/least respondent-level dataset from your Displayr Cloud Drive:

1. Under Data Sources, select + > Cloud Sources > Displayr Cloud Drive. 
2. Locate and select the respondent-level file created in Step 4 above. 
3. Click OK. 

Step 6 - Run the MaxDiff analysis

With the design and most/least variables in place, you are ready to run the MaxDiff:

1. In the Report tree, select + > Advanced Analysis > MaxDiff, then select either Hierarchical Bayes or Latent Class Analysis. 
2. In the max.diff Properties , select the following inputs: 
   • Design - select the md.design.widget from the design output created in Step 3 above. 

   • Add alternative labels - click this button to enter labels for each alternative starting in Cell A2. 

     Tip: Copy and paste the labels from Data Sources by selecting all of the alternative variables in the original Alida dataset, then right-click and select Copy Labels.

   • Version - select the Version variable from the respondent most/least dataset. 
   • Best selections - select the most1, most2, most3, etc. 
   • Least selections - select the least1, least2, least3, etc. 
3. Click Calculate.

Next

How to Perform an Advanced Analysis of Experimental Data (MaxDiff)

How to Export Data From Alida Sparq Platform (Formerly Known as Vision Critical)