Have an idea?

Visit Sawtooth Software Feedback to share your ideas on how we can improve our products.

CBC with alternative specific design and overlapping price levels

Have a CBC with alternative specific design.  Looking to compare sets of features (BASIC and PREMIUM).  There are 4 options on each task, 2 BASIC (in positions 1 and 2) and 2 PREMIUM (in positions 3 and 4).

The alternative specific design comes in because the prices for BASIC are different than the prices for PREMIUM.  Same with an additional subscription price.  So we will have 5 attributes but only showing each option with only 3 pieces of information: BASIC or PREMIUM, a base price, and a subscription price.

So far so good.  We start to run into a little trouble because a lot of the prices and subscriptions overlap between BASIC and PREMIUM.  Meaning that we will get a lot of BASIC options that cost more than PREMIUM options (the interplay between base price and subscription sometimes counterbalances but sometimes both cost more for the BASIC).  We don’t want to waste people’s time by providing obvious choices so we have put in prohibitions for some of these combinations – but not all.  By the same token in several cases, respondents will see 1 BASIC option being obviously cheaper than the second BASIC option – the choice between the 2 BASIC options presented to them is then easy but they still have to compare and choose versus the 2 PREMIUM options on the page.

I wanted to ask if having a few of the options presented be ‘obvious’ choices would detrimentally affect the utilities.  I wouldn’t think so, but wanted to confirm.  At the end of the day, want to have my simulator allow me to configure a BASIC and a PREMIUM product each with a base price and subscription price in the market and tell me share of each – vs. none.
asked Jan 30, 2014 by stevetlg.com (385 points)
retagged Jan 30, 2014 by Walter Williams

1 Answer

+1 vote
This is a very nice question.

First, I have to say there are some advanced power tricks (too complicated to get involved with here) that let you have your cake and eat it too: always make sure the premium model is more expensive than the base model, but accomplish that by not compromising the orthogonality of your experiment.  If you worked with an advanced consultant, you could accomplish that with some power tricks within our software.  I have nothing written on that I can share, if that is your follow-up question.  Sorry.  This topic has been discussed at "Turbo CBC" seminars in the past by David Lyon (the next Turbo CBC is going to happen in Amsterdam in October, BTW).

Next, the debate regarding having conjoint studies that occasionally show dominated concepts has been in play since the 1970s.  In the 1970s, they worried about orthogonal arrays for card-sort conjoint producing some cards that were inferior (logically) to other cards within the same experiment.  Researchers experimented with such orthogonal arrays versus other arrays (sometimes with prohibitions) that avoided dominated concepts (called Pareto efficient designs).  Green and Srinivasan reported in their 1978 article that multiple research projects investigating this were inconclusive: both approaches worked about equally well.  In other words, there didn't seem much harm in having occasional dominated concepts.

Pinnell and Huber looked into this for CBC studies in the 1990s and found very modest gains to be had by avoiding dominated concepts.  Paul Johnson implemented another experiment along these lines and presented it to the Sawtooth Software conference in 2013.  Same essential findings.
answered Jan 30, 2014 by Bryan Orme Platinum Sawtooth Software, Inc. (150,315 points)
Very much appreciate the insight here.  Given the business needs for this design, seems like a very workable approach.