The psychometrics of rating facial attractiveness using different response scales

Abstract

Perceiving facial attractiveness is an important behaviour across psychological science due to these judgments having real-world consequences. However, there is little consensus on the measurement of this behaviour, and practices differ widely. Research typically asks participants to provide ratings of attractiveness across a multitude of different response scales, with little consideration of the psychometric properties of these scales. Here, we make psychometric comparisons across nine different response scales. Specifically, we analysed the psychometric properties of a binary response, a 0–100 scale, a visual analogue scale, and a set of Likert scales (1–3, 1–5, 1–7, 1–8, 1–9, 1–10) as tools to measure attractiveness, calculating a range of commonly used statistics for each. While certain properties suggested researchers might choose to favour the 1–5, 1–7 and 1–8 scales, we generally found little evidence of an advantage for one scale over any other. Taken together, our investigation provides consideration of currently used techniques for measuring facial attractiveness and makes recommendations for researchers in this field.