Hello all,
Recently, I'm designing a factional factorial design using SAS. The experimental structure is attached as below:
General alternatives | Alt1 | Alt2 | Alt3 |
Specific alternatives | Unimodal | SM_multimodal | PT_multimodal |
Mode options | Car, Bike, Walking | Shared car – Walk, Shared E-bike – Walk Shared moped - Walk | BTM – Walk, Train – Walk, Bike – Train, Train – OV-fiets, Train – Shared E-bike, Train – Shared moped |
In-vehicle time | 70%, 100%, 130% | 70%, 100%, 130% | 70%, 100%, 130% |
Access/egress time | N/A | 5min, 10min, 15min | 5min, 10min, 15min |
Travel cost | 80%, 100%, 120% | 80%, 100%, 120% | 80%, 100%, 120% |
Generally, 3 alternatives * 4 attributes * 3 levels are applied here. There should be 1 attribute with 6 levels and 10 attributes will 3 levels. But not all mode options are available for the respondent. For example, a car will not be your option if you don't have a driver's license. In this case, the number of mode options levels is 2 not 3. Levels of this attribute will be extracted based on other questions in my survey. So it makes nonsense if I distribute car to a noe-car person based on the factional factorial result. Anyone have ideas what can I do?
Best,