Sensory difference tests for margarine: A comparison of R-Indices derived from ranking and A-Not A methods considering response bias and cognitive strategies

Sensory difference tests were performed between 6 margarine products, a standard vs 5 other products. Three testing protocols were used. The first protocol was simple ranking. The second protocol was the A-Not A method where a single standard was presented beforehand and which could be retasted during testing. The third protocol was the A-Not A method where all products were presented beforehand but could not be retasted during testing. R-Index values were computed for each protocol. Ranking gave the highest R-Index values while the A-Not A method, where only a single standard was presented prior to testing, gave the lowest R-Index values. R-Indices were calculated by averaging indices from individual judges and also by pooling data from all judges. Differences between these computations only occurred for the A-Not A method where all the products were presented prior to testing. Differences were explained in terms of the forced-choice nature of ranking, boundary variance, concept formation and differences in cognitive strategies involving tau and beta-criteria.