The authors develop a model which permits to analyze preference scores for paired comparisons: the binomial/Dirichlet model. This model requires less data collection than competing models , i.e. a smaller sample size and less data per respondent, to obtain the same level of parameter accuracy. Fitted to the preference scores provided by 44 product tests, the model shows a relatively good fit. Link to the article
BEMMAOR, A.C. and WAGNER, U. (2000). A Multiple-item Model of Paired Comparisons: Separating Chance from Latent Preference. Journal of Marketing Research, 37(4), pp. 514-524.