Pair Comparison, with and without Ties

Be your own placebo: A double paired preference test approach for establishing expected frequencies

“…Besides being valuable, because it adds information to the data (Gridgeman, 1959) a further argument for the 'No Preference' option is that without it, it can be difficult to interpret the data. Should half the consumers report a preference for product 'X' and half report a preference for product 'Y', no clear conclusion can be drawn if there were not a 'No Preference' option.…”

The paired preference test and the ‘No Preference’ option: was Odesky correct?

“…As support for the ÔNo PreferenceÕ option, Gridgeman (1959) argued that it is a valuable source of extra information. A further argument for the ÔNo PreferenceÕ option is that without it, it can be difficult to interpret the data.…”

Corroborating the 2-AFC and 2-AC Thurstonian models using both a model system and sparkling water

“…However, she did not use a specific Thurstonian model to calculate d 0 values but merely ignored or redistributed the ''no difference'' answers and used the available 2-AFC model (an easy and user-friendly d 0 calculation for the 2-AC method is shown in Appendix A of the present article). Gridgeman (1959) investigated the statistical power of the two methods and found that the 2-AC was slightly more powerful. However, he was not able to relate the two protocols more closely as he did not use a universal theory such as that based on Thurstone's ideas.…”