This use case examines how to set up your analysis in Advanced Exploration if you want to analyze data from a Multi-Select question, and view the data for each Multi-Select answer side-by-side.
The data for Multi-Select variables is comprised of 0 and 1:
- 0 means that this respondent did not select this answer choice
- 1 means that this respondent selected this answer choice
Since the data for Multi-Select variables is comprised solely of numbers, each of your Multi-Select variables can be found in your list of Measures.
Individual answers in a Multi-Select question are treated as different variables. As such, if your Multi-Select question has 5 answers, you will find 5 different variables in your list of Measures.
For this use case, we'll address the following request:
I want to compare my visitors’ level of interest with a number of different features.
Step 1: Select your Columns (X-Axis) and Rows (Y-Axis)
The Measure Names and Measure Values variables can be leveraged to analyze the data for multiple Measure variables at the same time.
- Measure Names (Dimension): Lists the names of all of the Measures for your survey
- Measure Values (Measure): Calculates the values for all of the Measures for your survey
The Measures Names and Measure Values variables can be found at the bottom of your list of Dimensions and Measures, respectively.
This will display each of your Measures in chart format, as well as their Sum values.
As well, a “Measure Values” card will appear below the Marks card through which you are able to specify how the value for each Measures is calculated (see Step 3).
Step 2: Specify the variables that want to analyze
By default, completing Step 1 will display data for all of your Measures. As such, the next step involves specifying which variables you want to analyze.
Completing Step 1 will automatically add the “Measure Names” variable as a Filter, which will list all of your Measures to the right of the chart.
Using the checkboxes provided, select the Multi-Select variables that you want to analyze:
Step 3: Specify how you want the data for the Multi-Select variables to be calculated
Completing Step 2 will filter the list of variables in the “Measure Values” card down to the ones that you selected.
Through this menu, you can specify how each selected Measure variables will be calculated by right-clicking each individual variable.
In this example, we want to examine the percentage of respondents who selected the different answers in our Multi-Select question. This will help us determine in which of the listed features they are most interested.
Since the data for our Multi-Select variables is comprised of 0 and 1, calculating the average for each variable will give us a value between 0 and 1 which, when multiplied by 100, will indicate the percentage of our respondents who selected this answer choice.
The scores for each of the Multi-Select answers will be between 0 and 1 which, when multiplied by 100, represents the percentage of the respondents who were shown this multi-select question that selected these answers.
Additional Use Cases