Premium
Conventional Analysis with Categorical Data from a Statistically Designed Experiment
Author(s) -
Goh T. N.
Publication year - 2017
Publication title -
quality and reliability engineering international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.913
H-Index - 62
eISSN - 1099-1638
pISSN - 0748-8017
DOI - 10.1002/qre.2069
Subject(s) - categorical variable , ordinal data , computer science , ordinal scale , software , parametric statistics , data mining , software quality , interpretation (philosophy) , econometrics , statistics , machine learning , mathematics , software development , programming language
Insights are offered on the interpretation of results of analysis of designed experiments with response expressed on a nominal or ordinal scale, in terms of formulation of a cause‐and‐effect mathematical model as well as the subsequent choice of factor settings for a future desired response. As generic design of experiments software packages are based on procedures of parametric statistics, the inherent limitations peculiar to the analysis of categorical data by such software packages are illustrated by a numerical example for the benefit of non‐statisticians among quality practitioners. Copyright © 2016 John Wiley & Sons, Ltd.