Overview of Nonparametric
Data Analysis Techniques

The objective of this course
is to acquaint students with the basic ideas, applicability, and methods of nonparametric data analysis.
After an introductory overview of fundamental concepts,
students will learn ten nonparametric analysis methods.

The course will be
laboratory-based. Use of the MINITAB™
statistical software package will be demonstrated using data from case studies.
Guidelines for interpreting MINITAB™ output analytically will be presented.
Students are encouraged to bring their own data analysis problems to the class
for discussion.

·
Review of hypothesis
testing

·
The comfortable cocoon
of normality

·
Detection
of non-normality – emerging from the cozy cocoon

·
Nonparametric tests

1
The one-sample
sign test

2
The one-sample Wilcoxson test

3
The Mann-Whitney test

4
The Kruskal-Wallis test

5 Rank correlation (Spearman’s r and Kendall’s t)

6 Cochran’s test for related observations

7
Mood's median
test

8
The Friedman test

9
The runs
test

10 Squared ranks test for variances

·
Summary

Work experience using basic
statistics and/or one basic statistics course (which included the topics of the
normal distribution and hypothesis testing) is recommended.

Conover, W. J. 1999. *Practical
Nonparametric Statistics*, 3^{rd} edition. New York, New York: John
Wiley & Sons, Incorporated.

Supplemental notes prepared
by the instructor.

Access to Web site prepared
by the instructor.

Sprent, P., and N.
C. Smeeton. 2001. *Applied Nonparametric Statistical
Methods*, 3^{rd} edition.
Boca Raton, Florida:
Chapman & Hall/CRC.

Edward J. Williams, Adjunct
Lecturer of Industrial and Systems Engineering, and of Management and
Information Science

This course is intended for
analysts, engineers, designers, and managers involved in the collection and
interpretation of potentially non-normal observational and/or experimental data.