Proponents of numerical conjoint measurement gen erally assume that the technique's goodness-of-fit mea sure will detect an inappropriate composition rule or the presence of random response error. In this paper a number of hypothetical and real preference rank order ings are analyzed using both axiomatic conjoint mea surement and numerical conjoint measurement to dem onstrate that this assumption is not warranted and may result in a distorted scaling.
No takes yet. Share an insight, caveat, or question.
Nickerson et al. (1984) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: