Abstract The latent nature of many phenomena we study in marketing necessitates the use of multi-item scales to adequately reflect construct richness. Ensuring that these scales are both valid and reliable provides a critical foundation for subsequent research. In pursuit of this goal, we propose a refined scale-development process model that builds on existing standards while emphasizing the role of theory for creating meaningful, valid, and novel scales. Further, we use a “methods-in-use” approach to systematically review scale-development papers published in premier marketing journals over the past 25 years to identify areas both where marketing scholars appropriately use established guidelines, but also where elements are executed with insufficient rigor. This leads to recommendations to help scholars avoid some common pitfalls we observed in reviewing and evaluating extant scale-development research. Finally, drawing on our framework in conjunction with observations of current scale development practices as well as recent methodological advances (e.g., the emergence of AI), we delineate best practices for creating high-quality scales.
Hulland et al. (2026) studied this question.