Examining the structure of visual analogue scales to capture motivation to eat in fasting and post-meal conditions

Abstract

The visual analogue scale (VAS) methodology for tracking hunger, fullness, desire to eat and prospective consumption attempts to capture conceptually distinct but related dimensions of motivation to eat. It is the most commonly used methodology to measure subjective motivation to eat in human appetite and energy balance research.

The current paper examined the underlying factor structure of the 4 motivation to eat VAS: 1) in 552 participants from 13 studies at the Human Appetite Research Unit (HARU) at the University of Leeds through exploratory factor analysis (EFA) and confirmatory factor analyses (CFA) in fasting and post-meal conditions; 2) in 151 participants of the multi-center DiOGenes study through CFA in fasting and post-meal conditions before and after weight loss.

EFA results indicated that >60 % of the variance between the VAS variables was explained by one underlying factor. The CFAs confirmed that the one-dimensional structure presented an overall good model fit. The 4 VAS questions presented high factor loadings. The one-dimensional structure also revealed high construct reliability and convergent validity across the 13 studies. A second analysis further confirmed a one-factor structure in fasting and post-meal conditions before and after weight loss. Measurement invariance testing was conducted across sex and fasted vs non-fasted conditions. Results indicated model invariance across sex at the configural, metric, and scalar levels, and partial metric invariance across conditions.

This current analysis indicates that hunger, fullness, desire to eat and prospective consumption VAS questions contribute to a single latent factor that should be used as a composite measure of the underlying process of motivation to eat. Additionally, this work suggests new methods should be developed to identify and measure different dimensions of motivation to eat states.