Top  Previous  Next

Show Effective Sample Size (ESS) When Weighting


When respondents are weighted differently (such as when weighting respondents within market simulations), the Effective Sample Size (ESS) becomes lower than the raw count of the sample size.  For example, even though you may have collected 300 respondents, if you weight respondents differently the precision of the results (the confidence interval) is negatively affected, as if effectively you have a smaller sample size, such as would result from 250 respondents.  The ESS is used in the standard error calculations when weighting respondents.  For more information, see: Effective Sample Size.


Tuning Factor (Exponent)


The market simulator lets you scale the utilities within the Share of Preference or Randomized First Choice options at the time the simulation is done.  This is accomplished by a parameter called the Exponent that you can set when preparing for simulations.  


The default value of the exponent is 1 (which multiplies all the utilities by 1 prior to calculating shares of preference).  The exponent can be used to adjust the sensitivity of the simulation results so that it more accurately reflects out-of-sample holdout choices, or actual market behavior.


A smaller exponent causes small shares to become larger, and large shares to become smaller — it has a "flattening" effect.  In the limit, with a very small exponent (near 0) every product receives the same share of preference.


A large exponent causes large shares to become larger, and small shares to become smaller — it has a "sharpening" effect.  In the limit, a very large exponent produces results like those of the First Choice option.

Page link: