Top  Previous  Next

A utility score is associated with each attribute level.  The higher the utility, the more preferred.  Utilities should only be compared within attributes, not between attributes.  In other words, the utility of a level for one attribute should NOT be compared to the utility for a level of a different attribute.  Utilities are additive across attributes, so the total utility of a product alternative is the sum of the utilities for its attribute levels (taking one and only one level from each attribute).


Due to the zero-centering of utilities within each attribute, even if all levels within the same attribute are considered excellent by respondents, some levels are made to be positive and others to be negative (to achieve zero-centering).  So, just because a level has a negative score doesn’t mean that respondents dislike that level (they just like it relatively less than the other levels within the same attribute).  The same logic applies that a positive utility does not necessarily mean that respondents like that level (they just relatively prefer it to the other levels within the same attribute).


For reporting purposes, the utilities per attribute are scaled to be normalized "zero-centered diffs".  Within each attribute, the utilities sum to 0 (so a negative score just means the level is relatively less preferred than other levels within the same attribute). For zero-centered diffs scaling, across attributes (for each respondent), the raw utilities have been multiplied by a constant so that the range of utilities for attributes averages 100. This rescaling procedure gives each respondent nearly equivalent weight when computing average utilities across the sample.


Even though utilities have been rescaled to zero-centered diffs for reporting purposes in Discover, it’s important to remember that the raw utilities are used in the market simulator to predict respondents’ choices.


Rather than reporting utilities to clients, we recommend showing the effect that changes from one level to another within attributes have upon product choice (share of preference) in the context of market simulations (i.e. sensitivity analysis).


Download Individual-Level Utilities


You can download the CBC utility scores for each respondent, for your attributes and levels, to a spreadsheet-compatible file by clicking Data | Summary | Download Data.  In addition to the general survey questions, you will find columns referring to the utility scores.


Some advanced users may wish to download the raw utilities (the utilities used for market simulation computations).  To download raw utilities, under Data | Advanced Settings you can specify that you want to download the raw utilities.  The file downloaded by clicking on Data | Summary | Download Data will now include the raw utilities.

Page link: https://www.sawtoothsoftware.com/help/discover/manual/index.html?utilities.html