Well, you know, discrete choice analysis of consumption bundles (options in your case) may require an example. Suppose N consumers in your experiment can buy a car and they can decide about such a purchase during 36 months (3 years ahead). You have the attributes of those two cars (options) BUT you should have the attributes of the status-quo option, I mean, the attributes of the used car each consumer owns. Other example, N consumers usually purchase one of two types of fresh fish every 2-week period and your study contemplates the decisions made by those N fish consumers during 72 periods (again, 3 years ahead). You have the attributes of those two types of fresh fish BUT you should have the attributes of the status-quo option: those of the (average quality) tuna containers each consumer can also purchase. Therefore, you must have at least 1 attribute level for the relevant status-quo option. Hope this helps!

Carlos R. Barrera-Chaupis

Thank you for the answer! Indeed, it is helpful.

I put all of my attributes also for the status quo option but it was 0. For instance, for my attribute donation, there were 3 levels such as 6 - 10 - 20 euros but the status quo option was 0 meaning they they would prefer not to choose other two options.

In this case, should I input also the status quo option for my analysis? I would like to be sure because putting status-quo to analysis will effect the coefficients and p-values of my attributes.

I agree with Cordula.

Besides the analytical part of DCE data, Status Quo options in the choice card design of DCE are particularly needed to account for protest responses. This is a different topic, but worth mentioning:

Villanueva, A. J., Glenk, K., & Rodríguez‐Entrena, M. (2017). Protest Responses and Willingness to Accept: Ecosystem Services Providers’ Preferences towards Incentive‐Based Schemes. Journal of Agricultural Economics, 68(3), 801-821.

Thank you Cordula Hinkes

Systematic inclusion of status quo alternatives can create problems beyond those described above (and possibly overcome by effect coding). I suggest the reading of a recent paper on the topic published in Public Choice 185(2020)1-19. Ask me or my co-authors for a copy if you do not have access to it.

Hi, how did you input the status quo into data for stata analysis? Now, I encounter the same question as you, look forward your help. Thanks

Cordula Hinkes