Brand coolness: development and validation of a short scale

Abstract

Purpose This study aims to present a shorter, more managerially useful, but psychometrically sound, scale for representing brand coolness than proposed by Warren et al. (2019). Design/methodology/approach The authors derive a 20-item scale based on the 37-item scale proposed by Warren et al. (2019) and examine its convergent, discriminant and predictive validity on data from one specified brand. Tests of hypotheses are performed on a random sample of 413 US adult consumers. Findings The new 20-item scale fits well the 10-factor confirmatory factor analysis model proposed by Warren et al. (2019). A multitrait–multimethod matrix analysis establishes convergent and discriminant validity. In addition, brand coolness adds unique predictive contributions explaining behavioral intentions and word of mouth over and above that provided by attitudes and self-brand connections toward the brand. Research limitations/implications This investigation examines only one brand (Nike). It would be desirable to test the generalizability of the model and measures of brand coolness on other brands and do so in multiple countries. Practical implications The new scale for brand coolness offers opportunities and insights to companies about how to measure and create brand coolness. Originality/value The original Warren et al.’s (2019) study used a sample for each analysis comprised of heterogenous brands and therefore could not make statements concerning the utility of the model for predicting any particular brand. This study demonstrates that the short scale can be validly applied to individual brands. The new scale also achieves construct validity using the stricter multitrait–multimethod analysis criteria, which Warren et al. (2019) were unable to do.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 11 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.