Selecting single items to measure doubly concrete constructs: A cautionary tale

Abstract

Single-item measures have recently become more en vogue due to studies arguing in favor of their psychometric properties vis-à-vis multi-item scales. However, their effective use requires (1) expert raters to designate the focal construct as being doubly concrete and (2) researchers to find a good single item to represent the construct. This study examines whether expert raters identify the doubly concrete nature of constructs that prior research presents as exemplary in this respect. Furthermore, the study compares the efficacy of a broad range of selection mechanisms based on expert judgment and statistical criteria for identifying the best item in a scale. The results show that expert raters do not share the commonly held belief that researchers can validly measure constructs such as attitude toward the ad, or brand, with single items. Further analyses show that neither rater assessments nor statistical criteria prove valuable regarding identifying an appropriate single item from a set of candidate items.