How to analyze "constant Sum" question? | XM Community
Skip to main content
Solved

How to analyze "constant Sum" question?

  • March 2, 2020
  • 3 replies
  • 129 views
Hoda
Hello everybody! 1. What is the difference between "constant sum" as a separate question type and also constant sum as a sub-type of "Matrix Table"? When do we use each of them? 2. What is the best way to analyze them after getting the data? Any ideas and help would be so much appreciated.

Best answer by Kate

The two different question types generate two different interfaces for users to enter information for you. Constant sum as a question type helps you determine how one variable effects the whole picture: ! While the Constant Sum Matrix Table helps you explore two variables effecting the whole: ! So your question of how to best analyze would be a question of first, what is your goal? What are you trying to learn from your customers, and how many variables are at play that you need to account for? From there the analysis should be guided by your research question!

    3 replies

    K
    Level 5 ●●●●●
    Forum|alt.badge.img+19
    • Kate
    • Level 5 ●●●●●
    • Answer
    • March 2, 2020
    The two different question types generate two different interfaces for users to enter information for you. Constant sum as a question type helps you determine how one variable effects the whole picture: ! While the Constant Sum Matrix Table helps you explore two variables effecting the whole: ! So your question of how to best analyze would be a question of first, what is your goal? What are you trying to learn from your customers, and how many variables are at play that you need to account for? From there the analysis should be guided by your research question!
    Hoda
    Hoda
    • Hoda
    • Author
    • March 4, 2020
    Thank you so much for the answers! Now I am thinking when we ask respondents to put amount of money in the boxes and when we ask them to allocate 100 points to each category? I know again it depends on the RQ, so my main goal is to study the willingness to pay (WTP) for some cultural services in agricultural landscape. I am gonna use some manipulated photos to illustrate some threats to the source of cultural services for visitors of that setting and explore how much they would like to pay to preserve the landscape and prevent those changes. Now I am wondering if it is better to ask them an open-ended questions about that amount or give them a hypothetical points (e.g. 100) to allocate to each service instead. In first way I will have real monetary valuation and in the second one only percentage. Right? Thanks again!
    K
    Level 5 ●●●●●
    Forum|alt.badge.img+19
    • Kate
    • Level 5 ●●●●●
    • March 5, 2020
    I personally like getting the real monetary value for willingness to pay questions. But, again it depends on the context of your question. If you really want to price a service that you want to offer some potential customers, getting the real monetary value is definitely the route you want to go. However, if you're trying to see what someone's sort of "investment portfolio" would look like, you want the other method. EG: You have X amount of money, and these 3 investments you can make. How much would you give to each investment? Then percentages are easier for your respondent to understand than adding/subtracting monetary amounts.
    Terms & ConditionsAccessibility statement