MaxDiff
Do you measure consumer preference or importance for brands,
concepts or attributes? Have you ever reviewed attribute ratings
data and wished there were more differentiation?
If so, MRSI can use Maximum Difference Scaling (MaxDiff) to help
you determine what attributes or items are most important or most
preferred among consumers with more clarity and ease of use than
traditional ratings.
Typically, MaxDiff estimates importance scores for approximately
15-40 items, although more could be accommodated.
Respondents are shown various subsets of items and asked which
is most and least important. Each respondent completes 10-15 groups
of questions where each group contains a different subset of items.
Responses to these questions are analyzed using Hierarchical Bayes
(HB) techniques to calculate importance scores at the individual
respondent level.
Research shows MaxDiff scores provide greater discrimination
among items and between respondents on items than standard rating
scales and results also parallel traditional marketing research
methods.
MaxDiff is similar to conjoint analysis, but it is:
- Applicable to a wider range of research situations
- Easier for both researchers and respondents
Plus, the clarity of your results is better using MaxDiff since
the technique more clearly identifies most and least important
items. This best-worst task is not only easy for respondents, but
it also eliminates scale bias, making MaxDiff ideal for
multicultural research and segmentation research.
MaxDiff's powerful design capabilities can also be applied in
developing sequential monadic designs for concept rotation. Designs
can effectively be created for virtually any number of items while
simultaneously balancing sample size by concept, concept pairs and
order of concept exposure.
Contact
MRSI for more information on MaxDiff today!
|