Non-Parametric Tests for Textile Ranking and Ordinal Data
topic
Kruskal-Wallis tests compare ranked distributions of textile properties across multiple groups without normality assumptions, Spearman rank correlation measures monotonic associations between textile property rankings, Friedman tests analyse randomised block designs with ordinal responses, and sign tests evaluate pairwise fabric handle and appearance preference comparisons from sensory panels.
Role
Provides valid statistical tests for textile data that violate the normality assumptions of parametric methods including sensory panel ratings, colour fastness grades, fabric handle assessments, and small sample measurements from non-normal distributions typical of extreme value properties like fabric defect counts and tear resistance.