Description
Analysis of Check-All-that-Apply (CATA) Data.
Description
Functions for analyzing check-all-that-apply (CATA) data from consumer and sensory tests. Cochran's Q test, McNemar's test, and Penalty-Lift analysis provided, as described in for CATA data analysis by Meyners, Castura & Carr (2013) <doi:10.1016/j.foodqual.2013.06.010>. Cluster analysis can be performed using b-cluster analysis. The quality of cluster analysis solutions can be evaluated using various measures. The methods related to b-cluster analysis are described in a manuscript by Castura, Meyners, Varela & Næs (2022) <doi:10.1016/j.foodqual.2022.104564>. Methods are adapted to product-related hedonic responses by Castura, Meyners, Pohjanheimo, Varela & Næs (2023) <doi:10.1111/joss.12860>.