128-qpman-research

Creating surveys arrow_right

Reports & analytics

Choice modeling arrow_right

Max Diff Scaling - Frequently Asked Questions

Can I set up an experimental design - where the total items are larger than the items presented in a single screen?

Yes. You can change the 'Maximum Attributes' and 'Attribute Per Task' to control the number of items displayed in a single screen.

Maximum Attributes: If you have numerous attributes, then you can limit the attributes displayed on the screen. For example, if you have 9 attributes and you set “6” in the drop-down, this means the respondent will only see 6 attributes out of the list of 9. If you set the drop-down values as “All”, this means all the attributes (i.e. 9) will be tested.

Attribute per task: Out of the selected maximum attributes we can select how many attributes we want to show per task. For example, if you set Attributes per task as “3” and Maximum Attributes as “All”, then you will have 3 cycles.

Survey Software Help Image

What kind of utility estimation model does the system use to calculate the share of preference?

The system uses the logit model.

How is the Preference Share calculated?

  1. Raw Utility (Best)

    U(b) = Ln(B(i) / (1 - B(i)))

  2. Raw Utility (Worst)

    U(w) = -Ln(W(i) / (1 - W(i)))

  3. Raw Utility

    Raw Utility [U(i)] = Average(U(b), U(w))

  4. Scaled

    U(Scaled)(i) = U(i) - Average(U)

  5. Transformed

    U(Transformed)(i) = EXP(U(Scaled)(i))

  6. Preference Share

    Preference Share = U(Transformed)(i) / SUM(U(Transformed))

Note: Back Button is not supported in the Max-Diff question.

License

This feature is available with the following license :

Research Edition

Was the article helpful?
Yes No

How can we improve it?

Submit

2 Questions

Can the positive and negative poles be reversed?

In edit mode, “least valuable” is on the left and “most valuable” is on the right. I can edit this, but I assume that the “minus pole” is still on the left and the “plus pole” is on the right for calculating the partial utility values, and that I cannot adjust this?

1 Answer
0
Individual-level MaxDiff utilities for segmentation

Hello QuestionPro team,

I want to use MaxDiff to create need-based segments from 20–30 items. I need individual-level utilities for SPSS, but it seems that QuestionPro only provides aggregate-level utilities via the logit model.

Is there a way to get individual-level utilities, or a recommended method within QuestionPro to achieve this?

Thank you!

0 Answer
0
You may also be interested in... MaxDiff Scaling - How It works. MaxDiff Scaling - Attribute Settings MaxDiff - Interpreting Results Anchored MaxDiff Analysis Related articles: Anchored MaxDiff Scaling – Making MaxDiff Bulletproof Anchored MaxDiff Scaling – Webinar Recap

Contact Us

Still have questions?

Webinar
Free Training
Support
certificate-icon Certification