The "None" Weight

Top  Previous  Next

If you are conducting simulations for a CBC study and the questionnaire contains choice tasks including a "None" option, then the option of None can be included in simulations (note that most ACBC studies result in a None weight as well).  However, the share of respondents predicted to choose None will only be correct if the number of products in the simulation is the same as the number of products in the CBC questionnaire.  This is another problematic outcome of the IIA rule and especially affects simulators built upon aggregate logit runs (which we strongly discourage).  With individual-level modeling under CBC/HB, we have seen that the share of "None" is approximately correct irrespective of the number of products used in the simulator (within a reasonable range of product alternatives).  See Orme (2017), "None Prediction in CBC: What Happens When We Simulate More or Fewer Alternatives than Were Shown in the Calibration Tasks?", working paper, Sawtooth Software.


If your conjoint analysis study has a None weight, the market simulator automatically includes the None as an option in the market simulation (on the My Scenario Settings tab, Simulation Settings ribbon group, the box is checked for Include none option (if present).


Previous versions of our market simulators allowed the user to specify a None weight.  If you want to adjust the strength of the None weight within this new simulator, you can adjust the utility of the None upward or downward via the Apply Share Adjustment feature within the External Effects program group on the ribbon bar.  

Page link: