The export options depend on whether you are using Traditional MaxDiff or Anchored MaxDiff:
Traditional MaxDiff Rescaling Export Options
Anchored MaxDiff Rescaling Export Options
Traditional MaxDiff Rescaling Export Options
For traditional MaxDiff studies (not anchored) there are three different scaling transformations that you can choose among to export your MaxDiff scores to a .CSV file:
Zero-Centered
These are the Zero-Centered Raw Scores where the scores average 0 and thus take on positive and negative weights. The raw (logit-scaled) scores that originally resulted from score estimation (either logit, latent class, or HB) and were shown to you in the report had the final item set to a utility of zero due to the dummy-coding procedure. To make it easier to make comparisons between groups or between people, for this export option we zero-centered the scores by subtracting the mean score from each raw score. These weights are on an raw logit interval scale, which does not support ratio operations. In other words, you cannot state that an item with a score of 2.0 is twice as important (or preferred) as an item with a score of 1.0.
We only recommend this rescaling procedure if the original logit scale needs to be preserved. However, due to differences in "scale factor" between respondents or groups of respondents, the Zero-Centered (Raw Scores) option can make comparing results between respondents or groups less robust.
Zero-Centered Interval
It is well known that raw logit-scaled scores involve a "Scale Factor" that is directly related to response error in the choices. The Zero-Centered Interval scaling transform places each respondent or group's scores on the same scale, where the difference between the best and worst scores is 100. This scaling transform has the benefit of removing differences in "scale factor" between respondents or groups of respondents such that the scores may be more correctly compared.
As with the Zero-Centered option, these scores are also zero-centered and thus take on positive and negative weights, but they take on much larger values in terms of absolute magnitude.
Probability Scaled
These individual-level item scores are positive values summing to 100 that reflect the likelihood of items being chosen within the questionnaire. Most researchers will probably use this scaling procedure, as it is easiest to interpret and present to others. This approach has the valuable property of ratio-scaling. That is to say, an item with a score of 20 is twice as important (or preferred) as an item with a score of 10. Click here for more details regarding the rescaling procedure.
Anchored MaxDiff Export Options
For Anchored MaxDiff studies, there are three different scaling transformations that you can choose among to export your MaxDiff scores to a .CSV file:
Raw
The anchor threshold is set to 0. The raw scores have logit scaling (the raw coefficients that result from using logit analysis and maximum likelihood estimation). Positive scores are associated with items judged above the anchor threshold; negative scores are associated with items judged below the anchor threshold.
We only recommend this rescaling procedure if the original logit scale needs to be preserved. However, due to differences in "scale factor" between respondents or groups of respondents, the Zero-Centered Raw Scores option can make comparing results between respondents or groups less robust.
Interval
If you have used Anchored MaxDiff, scores are displayed where the anchor threshold is set to 0. The zero-anchored interval scores are normalized to have a range of 100 utility points. Positive scores are associated with items judged above the anchor threshold; negative scores are associated with items judged below the anchor threshold. This scaling transform has the benefit of removing differences in "scale factor" between respondents or groups of respondents such that the scores may be more correctly compared.
Probability Scaled
The probability-scaled scores when using Anchored scaling are positive values where the Anchor is set to 100 and the largest possible value is the number of items shown to respondents in each set * 100. This approach has the valuable property of ratio-scaling. That is to say, an item with a score of 20 is twice as important (or preferred) as an item with a score of 10. See more details.