Any recommendations for easy to use and inexpensive software for conjoint analysis? I took graduate level stats more than a decade ago and would prefer something relatively easy to handle.


It depends on what type of conjoint analysis you're planning to do, and just what you're looking for from the software. If you're looking for easy-to-use, SPSS can analyze discrete choice experiments using the Cox regression function.

If you're looking for inexpensive (as well as more powerful) software, R can design and analyze discrete choice experiments. The package support.CEs provides some easy-to-use tools to design a paired experiment, although the options are somewhat limited. The package DoE.wrapper is a little more complicated, but provides far more design options. For analysis, the mlogit package can analyze simple multinomial logit as well as more complex probit and mixed logit models. It also comes with an invaluable set of tutorials. For very simple analyses, you could also try the conjoint package.

If you're looking for friendly AND powerful, you should look at NLogit, although it's a bit costly. Sawtooth is the commercial standard, but astronomically expensive.

Murtaza Haider has posted some helpful videos showing how to get started with discrete choice analysis in SPSS and Stata/R

  • $\begingroup$ To summarize DartPravteer's response: there are no cheap and easy options. A couple of other options are: Q (q-researchsoftware.com) is probably the easiest of the packages to use if creating latent class and mixed logit models (but it does neither design nor simulation) and there is a hosted conjoint service at Survey Analytics (surveyanalytics.com/conjoint), but I do not know much about it. $\endgroup$ – Tim Nov 16 '12 at 0:04

Not the answer you're looking for? Browse other questions tagged or ask your own question.