MaxDiff, or best-worst scaling, asks respondents to choose the most and least important items from repeated sets. It produces stronger discrimination than ordinary rating scales when the goal is to prioritise claims, needs, features or messages.
Read the full entry →Research Wiki · Methods Lab
MaxDiff and best-worst scaling
Best-worst choice designs that create sharper priority estimates than ratings when respondents must discriminate among items.
Methods family
Best-worst choice designs that create sharper priority estimates than ratings when respondents must discriminate among items.
Use methods only after defining the estimand, data structure, assumptions, validation plan and decision consequence.
Best-worst scaling is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Read the full entry →Case 1 object scaling is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Read the full entry →Case 2 profile scaling is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Read the full entry →Case 3 multiprofile scaling is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Read the full entry →Balanced incomplete block designs is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Read the full entry →Hierarchical Bayes MaxDiff is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Read the full entry →Aggregate logit MaxDiff is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Read the full entry →Latent-class MaxDiff is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Anchored MaxDiff is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Sparse MaxDiff is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Adaptive MaxDiff is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
MaxDiff versus rating scales is a decision comparison between MaxDiff and rating scales. The useful question is not which is universally better, but which approach fits the objective, evidence, population, timing and consequence of error.
MaxDiff versus ranking is a decision comparison between MaxDiff and ranking. The useful question is not which is universally better, but which approach fits the objective, evidence, population, timing and consequence of error.
MaxDiff sample-size planning is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
MaxDiff experimental design is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
MaxDiff score interpretation is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
MaxDiff respondent-level utilities is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
MaxDiff segmentation is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Maximum-difference scaling for claims is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
MaxDiff for feature prioritisation is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
MaxDiff for message testing is a method within maxdiff and best-worst scaling. Best-worst choice designs that create sharper priority estimates than ratings when respondents must discriminate among items. Its usefulness depends on data structure, assumptions, validation and whether the output answers the intended decision.
MaxDiff for need-state prioritisation is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Browse linked method entries
MaxDiff analysis
MaxDiff, or best-worst scaling, asks respondents to choose the most and least important items from repeated sets. It produces stronger discrimination than ordinary rating scales when the goal is to prioritise claims, needs, features or messages.
Open methodBest-worst scaling
Best-worst scaling is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Open methodCase 1 object scaling
Case 1 object scaling is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Open methodCase 2 profile scaling
Case 2 profile scaling is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Open methodCase 3 multiprofile scaling
Case 3 multiprofile scaling is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Open methodBalanced incomplete block designs
Balanced incomplete block designs is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Open methodHierarchical Bayes MaxDiff
Hierarchical Bayes MaxDiff is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Open methodAggregate logit MaxDiff
Aggregate logit MaxDiff is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Open methodLatent-class MaxDiff
Latent-class MaxDiff is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Open methodAnchored MaxDiff
Anchored MaxDiff is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Open methodSparse MaxDiff
Sparse MaxDiff is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Open methodAdaptive MaxDiff
Adaptive MaxDiff is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Open methodMaxDiff versus rating scales
MaxDiff versus rating scales is a decision comparison between MaxDiff and rating scales. The useful question is not which is universally better, but which approach fits the objective, evidence, population, timing and consequence of error.
Open methodMaxDiff versus ranking
MaxDiff versus ranking is a decision comparison between MaxDiff and ranking. The useful question is not which is universally better, but which approach fits the objective, evidence, population, timing and consequence of error.
Open methodMaxDiff sample-size planning
MaxDiff sample-size planning is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Open methodMaxDiff experimental design
MaxDiff experimental design is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Open methodMaxDiff score interpretation
MaxDiff score interpretation is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Open methodMaxDiff respondent-level utilities
MaxDiff respondent-level utilities is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Open methodMaxDiff segmentation
MaxDiff segmentation is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Open methodMaximum-difference scaling for claims
Maximum-difference scaling for claims is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Open methodMaxDiff for feature prioritisation
MaxDiff for feature prioritisation is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Open methodMaxDiff for message testing
MaxDiff for message testing is a method within maxdiff and best-worst scaling. Best-worst choice designs that create sharper priority estimates than ratings when respondents must discriminate among items. Its usefulness depends on data structure, assumptions, validation and whether the output answers the intended decision.
Open methodMaxDiff for need-state prioritisation
MaxDiff for need-state prioritisation is a statistical or analytical concept within maxdiff and best-worst scaling. It should be selected for the data-generating process and decision question rather than because software makes it available.
Open methodNo entries match this search.