within my current CBC study I found that for presenting realistic choice sets and still following the goal of the study, it might be necessary to use a combination of prohibitions and alternative-specific attributes.

Going through the Lighthouse help I found that in general such combinations should be avoided as they can lead to unexpected results.

Now I'm looking for some advice if the current setup of the study will yield valid results.

Can this be analyzed by looking at the design report? (see below)

Prohibition: Att1/Lev1 not in combination with Att2/Lev2,3

Alternative-Specific: Att5 only in combination with Att2/Lev2

Thank you very much in advance for any hints/explanations.

Kind regards,

Philipp

Build includes 120 respondents.

Total number of choices in each response category:

Category Number Percent

-----------------------------------------------------

1 412 34.33%

2 404 33.67%

3 384 32.00%

There are 1200 expanded tasks in total, or an average of 10.0 tasks per respondent.

Iter 1 Log-likelihood = -1313.14471 Chi Sq = 10.38007 RLH = 0.33478

Iter 2 Log-likelihood = -1312.87947 Chi Sq = 10.91055 RLH = 0.33485

Iter 3 Log-likelihood = -1312.86773 Chi Sq = 10.93404 RLH = 0.33486

Iter 4 Log-likelihood = -1312.86725 Chi Sq = 10.93500 RLH = 0.33486

Iter 5 Log-likelihood = -1312.86723 Chi Sq = 10.93504 RLH = 0.33486

Iter 6 Log-likelihood = -1312.86723 Chi Sq = 10.93504 RLH = 0.33486

*Converged

Std Err Attribute Level

1 0.10658 1 1 Immer Grün

2 0.06425 1 2 Stadtwerke Musterstadt

3 0.06318 1 3 Pink Strom

4 0.09970 2 1 Standard Strom ohne Label

5 0.06036 2 2 Öko Basis

6 0.06242 2 3 Öko Nachhaltig

7 0.05302 3 1 0 Prozent

8 0.05302 3 2 5 Prozent

9 0.05286 3 3 10 Prozent

10 0.05375 3 4 15 Prozent

11 0.04123 4 1 0 Prozent

12 0.04043 4 2 50 Prozent

13 0.04112 4 3 100 Prozent

14 0.08617 5 1 Grüner Strom

15 0.08674 5 2 TÜV Süd

16 0.08562 5 3 OK Power

In addition, I'd like to ask something about the D-efficiency that you mentioned. As I understood, those values can not be compared between different designs.

How about two designs that only differ in the number of random choice tasks? Is a direct comparison of the D-efficiencies between those two designs allowed?

Thank you very much!