Using matrix table score in loop and merge

Using matrix table score in loop and merge

I have set a loop and merge to assess the list of competitors on each attribute, but would like to ask the top 4 scored attributes of our list of competitors. Does anyone know how I could do this?

Best Answer

  • AkdashboardAkdashboard South Carolina Superuser ✭✭✭✭
    Accepted Answer

    Hi @Frankiesewell

    If I understand what you're trying doing, then there are about 1000 different ways to accomplish it (might be a slight embellishment :wink: ). Essentially you are looking at the top 4 rated questions for each brand, compared to your brand. For starters, make sure you have your numeric values correctly defined (may need to recode the question variables to make sure that "high" scores get a high number and "low" scores get a "low" number.) This seems like "duh" until you assume it is all correctly and you find out that the default numeric values are opposite what you intended (trust me, double check).

    1. If you have Vocalize: Make a simple bar chart widget for each brand that is the average score of the 10 categories, sorted from highest to lowest. Then select "Only Show" - "Top" [4]*

    2. If you don't have Vocalize and still want to use Qualtrics: In the "Report" section of your project, you can use the Gauge widget to average the scores for each question ( yes, this will be 10 gauges per brand, but it is pretty easy). I recommend using the advance Report section so that you have freedom to move the widgets in whichever way you like, see image below for a crude example). If this isn't an ongoing study, you can easily see which scores rank the highest and only keep the top 4 that are relevant. If this is ongoing, then it won't do you as much good.

    3. If you don't have Vocalize and don't care how it is displayed: Export to Excel and calculate the averages there (or really any tool that you are comfortable with... Excel is just a common one)

    Housekeeping note - Since you aren't using a very large number of potential questions [10q x (4 competitors + you) = 50q] and each set of 10 questions are identical, I wouldn't use Loop and Merge and instead would make each brand a separate block with branching logic in the survey flow. Then you can use the randomization feature at the block-level to display a random set of 4 questions. This is just personal preference, however I find this to be more straightforward when you go to analyze the results. Sometimes Loop and Merge data output can be confusing and require an (albeit simple) data dictionary.

Answers

  • AkdashboardAkdashboard South CarolinaCommunity Member Superuser ✭✭✭✭

    Hi @Frankiesewell

    I'd love to help you, but I am not sure what you are saying. Can you describe what you are trying to do a bit more?

  • FrankiesewellFrankiesewell UKCommunity Member Qubie ✭

    Hi @Akdashboard

    Thanks for trying to help. I am designing a brand tracker and will be asking our customers to rate us on 10 brand attributes. After this, I am displaying a list of direct competitors and using the selected to be rated on 4 of the 10 brand attributes. I have this set up as a loop and merge with the variable being a random selection of 4 of the 10 attributes.

    What I am trying to do is to take the 4 attributes that were scored highest and use these to get a comparison rating for each of our competitors that they are aware of. Does this make sense?

  • FrankiesewellFrankiesewell UKCommunity Member Qubie ✭

    Hi @Akdashboard

    Thanks again for the prompt response and for taking the time to pull out the suggestions. I think I will drop the loop and merge approach in favour of the simplified block-level route.

    Thanks!!!
    Frankie

Sign In to Comment