Quantification of Subjective Ratings Through Conjoint Measurement Analysis
Abstract
Conjoint measurement theory is examined through a prototype example in which a fighter aircraft is subjectively rated on two factors. As a first step, a multifactor ordinal scale is developed. This ordinal scale provides a meaningful measure of aircraft quality. Interval scales of aircraft quality are produced by the basic analysis of variance model and two conjoint measurement methods: delta scaling and the computer algorithm MONANOVA. These methods produce interval scales that differ by constant factors, as guaranteed by the theorem for additive conjoint measurement. The interval scale does not appear to be an improvement over the ordinal scale in the prototype example. There is no assurance that a specific conjoint measurement model can be used to improve the data. Major changes in the interval scales are caused by small perturbations in the rating matrix.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 08, 1983
- Accession Number
- ADA138810
Entities
People
- David E. Greene
Organizations
- University of Texas at Austin