UPCOMING WEBINAR: The Roadmap for Market Researchers in the Age of AI Many times, you will want to only include respondents in analyses that have a completed survey, or more specifically, have answered all of a set of questions. These respondents would have non-missing data for all the variables of interest, also known as complete cases. This article describes how to create a filter variable that only includes cases with complete data (ie, no missing values) across selected variables. You can then filter out the respondents who dropped out of your survey, or rather, cases where there is incomplete data.
The raw data table below shows an example of how the Complete filter variable (in the last column) is calculated based on the variables selected in the columns to the right of it:
You can create your variable in two different ways:
Requirements
A Displayr Document with a Data Set.
Method - Automatic
- In the Data Sources tree, select the specific variable set(s) that you would like to use in the filter.
- Click + > Filter > Complete Cases.
- A new filter variable will appear in the Data Sources tree that you can apply to your outputs and analyses.
Method - Using R
- Create an R variable, see How to Create a Custom R Variable.
- In the object inspector, ensure Usable as a filter is checked.
-
Paste in and modify the following code in the R Code editor:
#make a table of all of the variables to identify complete cases
#replace the variable set names within data.frame() with a comma separated
#list of your own, you can click on the variable set in the Data Sources tree
#to insert it into the code
mytab = data.frame(Age,`D2 - Income`,Gender,`Q3 - Preferred cola`)
#calculate complete cases across each row in the table
complete.cases(mytab)
Next
How to Impute Missing Data in Displayr
How to Check for Missing Data Using Plot of Patterns