No One Optimal Way to Measure People’s Attitudes? Preferred Length of Scales in Advertising Research

Abstract

Rating scales are commonly used to measure constructs that are not directly observable, such as attitudes. The extant research in the literature has extended the general understanding of attitude measurement. Yet the optimal number of scale points remains elusive. The current study aims to discuss this topic in an advertising research context. I conduct a comprehensive review of the literature on the effect of scale points from 1924 to 2021 and summarize empirical evidence as well as major arguments over time. I also perform an analysis of rating scales measuring attitude toward the ad that appeared in the Journal of Advertising from 2011 to 2020 and explore how and why researchers have used some rating scale points. I present the key issues and propose several practical recommendations for future research.

Disclosure Statement

No potential conflicts of interest were reported by the author.

Notes

1 From the 1940s to the 1960s, researchers mainly focused on developing ways to evaluate attitudes and identifying the measures associated with cognitive or affective indicators while assessing their reliability, consistency, and susceptibility to biases (Eagly and Chaiken 1993). This led me to conduct a systematic review of attitude measurement in scholarly articles related to advertising and marketing published after the 1970s. I initially compared the measurement practices used in related fields with a 10-year interval and then performed a detailed analysis of the methodology employed in Journal of Advertising articles published from 2011 to 2020. Here I should note that the current methodological review did not aim to explore the evolution of measurement practices over time but rather concentrated on comparing the frequency of practices across various chronological phases.