The Contributions of Questionnaire Length, Format, and Type of Score to Response Inconsistency

Paired Comparisons Intransitivity: is it Relatively Stable Over Time?

“…Although intransitivity has been useful in determining subjects' inconsistency in responding to paired comparisons instruments, the present research supports previous studies (Ace and Dawis, 1972;Hendel, 1975) which indicated that intransitivity generalizes across instrument length, format and content and provides further support for response intransitivity as an individual differences variable.…”

“…Theoretical articles have examined intransitivity as an index of the validity of an individual's preferential system (Benson, 1958), as a function of the characteristics of the stimulus objects (Coombs, 1958), and as a reflection of the probabilisitic nature of choice behavior (Tversky, 1969), among others. Empirical studies have examined questions such as the extent to which intransitivity generalizes across instrument length and format (Ace and Dawis, 1972), the extent to which intransitivity generalizes across instrument content (Hendel, 1975), and the usefulness of the total circular triad score in differentiating among individuals with respect to reliability of measurement (Hendel and Weiss, 1970).…”

Type of Content, Type of Score, and Response Inconsistency in Comparison Measures of Preference

“…To determine if response inconsistency was not simply an artifact of the psychometric instrument itself, Ace and Dawis (1972) examined three instrument characteristics, length (number of items), item format, and type of score, as factors in a 3-way completely crossed analysis of variance experiment, with 13 observations per cell. There were two levels of length (instruments constructed from 13 stimulus statements vs. 7 stimulus statements, the latter being a subset of the former), two levels of format (paired-comparisons vs. triad rank order) and three levels of type of score (upper triangle circular triad score vs. lower triangle circular triad score vs. inconsistency score from a comparison of upper and lower triangles).…”

“…Inventory 2 was constructed from the same set of seven stimulus statements about work needs used in the earlier study (Ace and Dawis, 1972). Procedure Three preference inventories were developed, representing three different types of content.…”

“…The three types of scores used in the previous study (Ace and Dawis, 1972) were calculated from the constructed paired comparisons matrices. The respondents were asked to select the most and the least preferred stimulus in each triad, thus effectively rank-ordering the three stimuli.…”