Measuring consumer response to complex perception: Scaling vs. categorization task

Scaling is commonly used with consumers for measuring complex perceptions (e.g. creaminess, refreshing) induced by food. Since this task is prone to different psychophysical biases in scale usage that increase the inter-individual variability, the categorization task (CT) may be a more simple and discriminating alternative to scaling. In the present study the expected food naturalness of a packaged prepared food was evaluated by 120 consumers based on 8 different packaging pictures. Each consumer assessed 4 out of 8 pictures. First they categorized each picture as “Not at all natural” or “Extremely natural” while recording their response time. Then they rated naturalness using a line scale where “Not at all natural” or “Extremely natural” were the anchors. CT binary responses were transformed into continuous data using response time and compared to scaling data in terms of agreement and discriminative power. As key findings, CT results were in agreement with scaling results but discrimination power was greater with scaling than CT. As perspective, further research is required to determine whether the simplicity of CT may lead to more discrimination compared to scaling for specific consumer targets such as older population.

► We explore categorization task as a more discriminative and simple alternative to scaling for assessing naturalness with one hundred and twenty consumers.
► The impact of 8 food package artworks on expected food naturalness is measured using both methods.
► Scaling and categorization task provide results in agreement but discrimination power is greater with scaling than with categorization task.