Follow-up Question after MaxDiff | XM Community
Skip to main content
Michael_Cooksey
Administrator
24 days ago
Solved

Follow-up Question after MaxDiff

J
Forum|alt.badge.img+2
Is it possible to add a piped text follow-up question after a maxdiff exercise (with the piped text being their top option)? For example, after respondents go through a maxdiff exercise of 10 questions for "most important feature", would it be possible to ask "Please explain why [piped text] is the most important feature for you". I know there is the ability to have an anchored maxdiff question as respondents go through the series of questions, but I imagine there would be a high risk of respondent drop off having to answer why they chose their option 10 different times (we're only interested in the reason for the ultimate winner)

Best answer by Kate

Ah I see. So you would have to somehow indicate what the top choice was in the embedded data, then you could pipe that in. So long as you have a limited number of branches, you might be able to work out in the survey flow. If Q1 =1 AND Q2 =1 then Embedded data Selected Choice= A But if you have more than 2 exposures to the selection, the survey flow is unable to use nested logic.
View original

11 replies

K
Level 5 ●●●●●
Forum|alt.badge.img+19
  • Kate
  • Level 5 ●●●●●
  • 543 replies
  • October 30, 2019
Looks like a pretty straight forward pipe to me. Are you having issue with something in particular? !
J
Forum|alt.badge.img+2
  • jmejia
  • Author
  • 8 replies
  • October 30, 2019

No, that looks like it would only carry over the top choice for the first question of a maxdiff not the entire series of questions (e.g. testing 16 different features, see 4 features at a time for a series of 8 questions, selecting top choice and bottom choice each time). I would also be using the built-in maxdiff platform from qualtrics.

K
Level 5 ●●●●●
Forum|alt.badge.img+19
  • Kate
  • Level 5 ●●●●●
  • 543 replies
  • Answer
  • October 30, 2019
Ah I see. So you would have to somehow indicate what the top choice was in the embedded data, then you could pipe that in. So long as you have a limited number of branches, you might be able to work out in the survey flow. If Q1 =1 AND Q2 =1 then Embedded data Selected Choice= A But if you have more than 2 exposures to the selection, the survey flow is unable to use nested logic.
J
1 person likes this
  • J
    jmejia
TomG
Level 8 ●●●●●●●●
Forum|alt.badge.img+27
  • TomG
  • Level 8 ●●●●●●●●
  • 5926 replies
  • October 30, 2019
@jmejia, Two possible options: 1. Pipe all the "most" choices into JavaScript and count up the occurences to find the top choice. There's a chance there will be a tie for the top choice, so you'd have to decide how to handle that. 2. Have a follow-up question that asks for the top choice with all 16 features as choices. Add display logic to each choice to only display if it was picked as "most" in one of the maxdiff questions. Then pipe the answer into your why open end question.
J
1 person likes this
  • J
    jmejia
J
Forum|alt.badge.img+2
  • jmejia
  • Author
  • 8 replies
  • October 31, 2019
Thanks @Kate! > @Kate said: > Ah I see. > > So you would have to somehow indicate what the top choice was in the embedded data, then you could pipe that in. So long as you have a limited number of branches, you might be able to work out in the survey flow. > > If Q1 =1 AND Q2 =1 then Embedded data Selected Choice= A > > But if you have more than 2 exposures to the selection, the survey flow is unable to use nested logic.
J
Forum|alt.badge.img+2
  • jmejia
  • Author
  • 8 replies
  • October 31, 2019
Thanks @TomG! I will look into these options. > @TomG said: > @jmejia, > > Two possible options: > 1. Pipe all the "most" choices into JavaScript and count up the occurences to find the top choice. There's a chance there will be a tie for the top choice, so you'd have to decide how to handle that. > 2. Have a follow-up question that asks for the top choice with all 16 features as choices. Add display logic to each choice to only display if it was picked as "most" in one of the maxdiff questions. Then pipe the answer into your why open end question.
J
Forum|alt.badge.img+2
  • jmejia
  • Author
  • 8 replies
  • October 31, 2019
Update: After a little more digging, I found that including skip, display and branch logic is currently not supported with the Quatrics Maxdiff PX platform. The above options may only work if programming a maxdiff manually. Screenshot of Qtip and link to page I found the info below. ! https://www.qualtrics.com/support/conjoint-project/survey-tab-px/building-additional-survey-content-px/
A
Forum|alt.badge.img+1
  • AbeB
  • 3 replies
  • February 29, 2024

Has this feature been added to Qualtrics yet? Seems like a pretty big hole in the MaxDiff functionality which Sawtooth satisfies.

lizb311
Level 3 ●●●
Forum|alt.badge.img+20
  • lizb311
  • Level 3 ●●●
  • 119 replies
  • March 1, 2024

@AbeB I totally agree with you there. Unfortunately, it looks like it’s still not quite possible in Qualtrics if you’re using the built-in functionality (source).

 

A
1 person likes this
  • A
    AbeB
A
Forum|alt.badge.img+1
  • AbeB
  • 3 replies
  • March 4, 2024

@lizb311 , is the implication then that we’d need to manually code a MaxDiff into Qualtrics? That seems incredibly unwieldy logistically. 

lizb311
Level 3 ●●●
Forum|alt.badge.img+20
  • lizb311
  • Level 3 ●●●
  • 119 replies
  • March 4, 2024

@AbeB Unfortunately, yes. 😔 With the built-in MaxDiff project type, I have tried a few ways of getting at that info (adding embedded data to the Survey Flow, with Javascript in the survey header, with JS in a question that displays after the MaxDiff block, …) but haven’t had any luck getting it to tell me what people actually selected in that MaxDiff block yet. If anyone else has found a good solution, I’d love to learn! 

Leave a Reply


Terms & Conditions