UPCOMING WEBINAR: Build it once. Use it again: Create reusable charts with AI In some instances, a nominal (single-select) question may be coded as binary variables where each category is stored as a separate binary variable. In these cases, the data is mutually exclusive, meaning respondents selected one and only one answer. You will often want to convert these binary variables into a single nominal variable, as this makes it much easier to work with in terms of building crosstabs/banners, as well as using in other analyses. Other times, you may need to condense mutually exclusive data across variables for data manipulation purposes.
This article describes how to use the Convert To > Nominal feature to take a binary variable set with mutually exclusive data:
And create a new categorical (Nominal) variable with categories based on the labels of the binary variables:
Requirements
A dataset loaded in Displayr containing a binary variable set that has mutually exclusive data (i.e., respondents answered only one and only one option from the binary variables).
Method
- Under Data Sources, select all of the binary variables containing the mutually exclusive data.
- Hover over one of the selected variables and select + > Convert To > Nominal. A new nominal variable will be created.
- OPTIONAL: Select the new variable and click the Labels button in the properties to modify the category labels as needed.
Additional Notes
There may be times when the data is expected to be mutually exclusive, but there are actually cases where responses may have selected more than one option. This may be due to flawed survey programming, where multiple responses were allowed where only a single response should have been permitted.
When using the above method in these situations (or when trying this with a true multi-response variable set), you'll receive a warning that the variables are not mutually exclusive:
If you click Yes, the process will run, however, since the function loops through the variables, it will only capture the last variable selected. So it is important to ensure that your data is mutually exclusive before applying this process.
When the binary data being converted to nominal contains all missing values and/or there are respondents who have not selected any of the options in the multi-response question, you'll receive the following alert when doing the conversion:
You can click Yes to proceed with the conversion, however, these respondents will be coded as missing values in the new nominal variable and will therefore be excluded from the base.