Abstract:

Data obtained from many forced choice procedures such as the 2- and 3-alternative forced choice (2-AFC and 3-AFC), the duo-trio and the triangular methods, are often tested using the binomial distribution. In some applications of either difference or preference testing it is desirable to include a no difference or no preference option. One example occurs in claims support where there is a preference to include this option, but there are many other such applications in product testing for product development or quality assurance. Note that in this report we refer only to no difference counts although the ideas we present apply equally well to no preference counts. Also note that we refer to a two-alternative task with a no difference option as the 2-AC (2-alternative choice) in which the instruction is to choose the item of two (A and B) with the greatest or least amount of some attribute, or to indicate no difference.