Hi everyone,For our study we have six conditions, I already used embedded data en made them be presented evenly. Unfortunatly, when people don't complete the survey, the evenly presented might not be true. Since we will need an Anova test at the end, is there an easy way to check how it's trully spread in the completed data? Besides ofcourse just handcounting them, but that is a lot of work and more change to make mistakes yourself as well.I hope anyone has an idea for this. Thanks in advance!

FloorG Is the "Group" variable an embedded data field that you set in your Survey Flow in Qualtrics? If so, go back into your Survey Flow and change the type of the embedded data by clicking on "Options." Choose something like "Number Set" as the type, and then save it. Here's more information on the different types of embedded data options. That should allow you to find discrete values, counts, and totals in your Reports. You can change the embedded data options even after your survey has gone live without worrying about a negative impact on your respondents. They won't see the change; it's really more of a back end thing. Then once you get your embedded data set up properly, add a filter to ensure you're only looking at people who completed the survey.

What are you going to be using to analyze the data? And how exactly are you defining a "complete" response?You could create a new filter in your Qualtrics Reports tab that looks only at people with Finished = TRUE or Progress = 100. Then you would see how many responses are left in your data set after filtering, and if you have an embedded data variable that tells you the condition, then you would be able to see how many respondents fell into each condition.If you're using Excel, it could be as simple as a Pivot Table. Again, using Finished = TRUE or Progress = 100 as a filter, and a count of your condition variable. If you don't want to use Finished or Progress variables for whatever reason (e.g., because somebody could finish the survey and have Finished = TRUE but still leave some answers blank along the way), then try creating your own variable that checks to see if the response is "complete," however you want to define that. And if you don't have an embedded data variable that tells you which condition the respondent fell into, then you should definitely think about setting or creating one.

Easy check for evenly presented embedded data.

Easy check for evenly presented embedded data. | Experience Community

Sign up

Login to the Community

Scanning file for viruses.

This file cannot be downloaded