Conjoint analysis was traditionally carried out with some form of multiple regression. Nowadays the use of Hierarchical Bayesian analysis has become widespread as the golden standard, enabling fairly robust statistical models of individual respondent decision behavior to be developed.

Hierarchical Bayes: benefits & limitations

  • Can provide estimates of individual part worths based on relatively sparse data for each individual
  • Can take quite some time to estimate part worth utilities

HB: when to use it?

HB can be used whenever you want to model responses/preferences on the individual level instead of on (semi-) aggregate level. This applies both for Conjoint as well as for MaxDiff responses.