I have a 20 row matrix with 3 scale points (Yes/No,Don't know). The final row in the matrix is an 'Other' row. I would like to force users to select a Y/N/DK response for every row except 'Other' if that row does not apply to them. Using the native 'Force response' would require them to respond to the 'Other' row so I'm trying to use custom validation. I'd like to use a math operation to ensure that the sum of Yes (Count)+No (Count)+DK (Count) must equal 19. I can use these counts within the custom validation menu, but I am not sure how to reference these counts within a math operation. Can someone help with this? Or am I missing another obvious way to do this easily? Thank you!
Page 1 / 1
@Nadaly I'm curious what the motivation is for such a long matrix -- that's going to bring up some (potential) accessibility issues and it's a long question for the respondent to pay attention to.
Perhaps break it up into a few smaller matrices, and ask the "other" question separately (e.g., "would you like to suggest another area that is not accounted for in the previous question(s)?)
Perhaps break it up into a few smaller matrices, and ask the "other" question separately (e.g., "would you like to suggest another area that is not accounted for in the previous question(s)?)
@AdamK12 - Duly noted regarding the long matrix and potential accessibility issues. Valid point. But it's a fairly simple list of tools where "Yes" selections then drive a Loop & Merge block of follow-on questions, so I think we want to keep it intact. But if no one else responds, I may have to remove the "other" row. Thank you!
You can pipe the counts into right side of the validation comparison. So, your custom validation becomes like the following where the right side is math operation:
Yes count = 19 - piped No count - piped Don't Know count
Yes count = 19 - piped No count - piped Don't Know count
@TomG - YES! That's exactly what I'm describing! But I don't know how to specifically call on those counts or where to get that information in order to write the math operation formula more specifically. I know how to use other embedded data, like selected choices for row 1 would be ${q://QID10/ChoiceGroup/SelectedChoicesForAnswer/1} in this case, but what is the Qualtrics embedded data identifier for counts by scale point in a matrix?
> @Nadaly said:
> @TomG - YES! That's exactly what I'm describing! But I don't know how to specifically call on those counts or where to get that information in order to write the math operation formula more specifically. I know how to use other embedded data, like selected choices for row 1 would be ${q://QID10/ChoiceGroup/SelectedChoicesForAnswer/1} in this case, but what is the Qualtrics embedded data identifier for counts by scale point in a matrix?
It drives me crazy that Qualtrics doesn't expose all the piped fields through the interface. There is a Product Idea along these lines...please vote for it.
Anyway, I think your validation should be something like:
Yes (Count) >= $e{ 19 - q://QID10/SelectedAnswerCount/2 - q://QID10/SelectedAnswerCount/3 }
> @TomG - YES! That's exactly what I'm describing! But I don't know how to specifically call on those counts or where to get that information in order to write the math operation formula more specifically. I know how to use other embedded data, like selected choices for row 1 would be ${q://QID10/ChoiceGroup/SelectedChoicesForAnswer/1} in this case, but what is the Qualtrics embedded data identifier for counts by scale point in a matrix?
It drives me crazy that Qualtrics doesn't expose all the piped fields through the interface. There is a Product Idea along these lines...please vote for it.
Anyway, I think your validation should be something like:
Yes (Count) >= $e{ 19 - q://QID10/SelectedAnswerCount/2 - q://QID10/SelectedAnswerCount/3 }
Leave a Reply
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.