Need help using question logic and displaying question data from a dataset | XM Community
Solved

Need help using question logic and displaying question data from a dataset

  • 26 June 2018
  • 7 replies
  • 521 views
M
Hello,
I am trying to figure out how to use some advanced features and cannot figure out the best way to go about implementing what I need this survey to do after having experimented with Qualtrics and read some of the documentation. I have an idea about how to go about things, but want to query the community first.

I have a survey that asks a group of participants what their occupation is, then they select from a list of several answers. Depending on what answer they select they will receive a custom set of questions for that occupation. The questions they receive are tasks from another data set that the occupation routinely performs. For example, someone would select “administrator” and they would receive a custom survey that features the tasks associated with that occupation. Here is where things get complicated. For each task I have two questions I want participants to rate. So if they see the task “process paperwork” I want them to rate that task on two different five point scales. The same scales and questions are the same for every task, for every occupation. It is just a matter of showing each occupation about 10 tasks and using the same two ratings per question over and over. I have a dataset of about 50 tasks for each occupation I am interested in, so it would be nice to show each occupation 10 tasks, and the next time an administrator comes along they see a different 10 tasks from the total of 50 tasks I have for that occupation. Is there a way to do all this?

Thank you kindly for your thoughts on this and helping me to understand the best way to go about implementing this survey.
Best,
Matt
icon

Best answer by TomG 26 June 2018, 22:23

View original

7 replies

w.patrick.gale
Rank
Userlevel 5
Badge +13
@matthias Where are you in building the survey (progress)? I suggest tackling one thing at a time.
M
I'm happy to do it one thing at a time, once I actually know how to go about implementing each thing. I'm at the point in the survey where I need to figure out to do what I mentioned above 😀.

Basically, I know I need to use the display logic. But there has to be an easier way to do this than doing it for every question. I would imagine the slow and long process to doing this is creating a display logic for each occupation the participant selects, and then showing them custom questions. I just don't know how to have the same measures for every question and have the question text change based on an excel file I have. If that makes any sense. I'd rather not do all of this "by hand".
w.patrick.gale
Rank
Userlevel 5
Badge +13
@matthias For multiple tasks you might want to consider a matrix grid instead of single questions for each response so you only setup the display logic for one matrix set instead of individual questions. Then you would probably want to use the loop and merge function and some jQuery possibly for the question text change.
TomG
Rank
Userlevel 7
Badge +27
@matthias,

You want to use loop&merge and piping. Basically, you can set up one loop&merge block for each occupation that would include your two questions. You would loop over the tasks and pipe the current task into the questions. You can randomly loop over 10 of the 50 tasks for each occupation.
M
Ah! Thanks @"w.patrick.gale" and @TomG , that sounds like what I need to use to get the job done. I'll study up on loop and merge and see if I run into any issues. I'll post back if I run into a problem, but it looks straightforward after skimming the documentation Tom linked. Thanks again for the help!
M
Bit of and update:

Things are going well. I set up the loop and merge and have all of my "tasks" in per question. Everything is working as I had hoped.

Now I 'm not sure how to tackle the second part of this problem. For reasons of time it will take to complete the survey and trying to avoid fatigure, I do not want to show respondants more than 10 tasks they are rating. So, I would like to show them a smaller sub set of the larger task data set I have. If I have 20 tasks I need to rate I would like to show two participants a set of 10 tasks each so that all 20 are rated atleast once. I see I can randomize how these data are being selected and shown in a question. But I do not want some tasks rated twice and others not at all. Maybe this is where the Javascript comes in. But I'm not sure how to go about having a participant see the first set of 10 tasks, the second participant the second set of 10 tasks, and so on till all of the tasks are rated, and then starting that cycle all over again once the data set has been passed through once. That cycle would keep happening either until I hit the numebrs I'd like, or I run out of prospective participants.

Would deeply appriciate any thoughts and pointers on this issue.
TomG
Rank
Userlevel 7
Badge +27
Create an MC question before each loop with your tasks as choices. Use Advanced Randomization to randomly display 10 choices and check “Evenly Present”. Make sure it is NOT forced and use JS to hide it (preferably put it on the same page as other questions so the JS doesn’t have to click the Next button). Set up your ensuing loop to loop based on the displayed choices in your MC question. Turn off randomization in your loop setup. Add additional loop & merge fields if needed.

Leave a Reply


Terms & Conditions