Rating scale data need not be manipulated to satisfy the assumptions of the analysis of variance, but formed into an incidence table classified by the ratings and other factors, and compacted over factors to focus on those which are important. References are given to methods for formal statistical analysis of these tables.
No takes yet. Share an insight, caveat, or question.
L. P. Lefkovitch (1991) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: