Week 1

Environmental Course code: GST211

Sustainability Topic: Environmental Studies

Course Name

Introductory
Statistics I

Topic: An Introduction to Non-parametric Statistics

Week 12

© Copyright 2021 MIVA Open University All Rights Reserved

Week 1

Introductory Topic: Introduction to Statistics and Data

Statistics I

In Summary

WHITE NOISE WHITE NOISE + LINEAR TREND

+ +
This week, you explored the field of non-parametric +
+
+ ++
statistics, which is a branch of statistics that does not
+
−
−
assume a specific probability distribution for the data being −
− −

−−
analysed. You learned that non-parametric statistics provide +− + +-+ − − | + − + − − + − + − − − − + − − + + ++++++
r=1 2 3 4 5 6 7 8 9 r=1 2 3 4
an alternative approach to classical parametric statistics HF OSCILLATION
+ +
LF OSCILLATION
+ +
+ +
when the assumptions of parametric tests are not met. + + + + +
+ + +
+
+
Also, you learned that non-parametric tests are used −
−
−
when comparing groups of data, testing for differences
−
− − −
− −
−
between variables, or testing for correlation. In addition, you
− −
+− + +-+ − − | + − + − − + − + +++ − − − − − − − + + +++
r=1 2 3 4 5 6 7 8 9 10 11 12 13 r=1 2 3

explored examples of non-parametric tests which include

the Wilcoxon rank-sum test, the one-sample sign test, the
0.1
Kruskal-Wallis test, the Kolmogorov-Smirnov test, and the
runs test. Overall, you learned that non-parametric statistics 0.8

offer valuable tools for analysing and testing data without

0.6
relying on strict distributional assumptions.

0.4

0.2

Bullet point summary

• Non-parametric statistics do not assume a specific tests.

probability distribution for the data being analysed. • Examples of non-parametric tests include the Wilcoxon
• Non-parametric tests are used when data does rank-sum test, one-sample sign test, Kruskal-Wallis
not meet the assumptions of parametric tests, test, Kolmogorov-Smirnov test, and the runs test.
such as non-normal distributions or unknown
population parameters. 1.2

• Non-parametric tests are suitable for nominal or

1.2

1.2

ordinal scale data and situations where parametric 1.2

1.2

assumptions are violated. 1.2

1.2
• Non-parametric tests use ranks of data observations 1.2

instead of the actual values.

0.1

0.8

• Non-parametric tests are considered more robust 0.6

0.4
but may have less power compared to parametric 0.2
0
0.1 0.2 0.3 0.4 0.5 0.5 0.6 0.7 0.8

Key references:
© Copyright 2021 MIVA Open University All Rights Reserved

Kolassa, J. E. (2017). Nonparametric Statistics: A Step-by-Step Approach. Wiley

Gibbons, J. D., Chakraborti, S. (2011). Nonparametric Statistical Inference (5th ed.). CRC Press
Week 1

Introductory Topic: Introduction

Course code: GST211to Statistics and Data
Statistics I Topic: Environmental Studies

Case study

Title: Comparison of Customer Satisfaction Objective: The objective of this study is to compare
Ratings Using Non-parametric Tests the customer satisfaction ratings of two different
companies in the retail industry using non-parametric
Introduction: Customer satisfaction is a tests.
critical factor in the success of any business.
To effectively measure and compare customer Methods: The method involves collecting customer
satisfaction ratings, companies often utilise satisfaction ratings from two different companies,
statistical methods. This case study focuses preparing the data for analysis, formulating
on the application of non-parametric tests hypotheses, selecting an appropriate non-parametric
to compare customer satisfaction ratings test, conducting the data analysis, interpreting
between two or more groups. the results, and drawing conclusions based on the
comparison of customer satisfaction ratings.

Questions to ponder

• What are the advantages and

disadvantages of using non-
parametric tests compared to
parametric tests?
• How can non-parametric
tests be applied in real- SKILLS AND COMPETENCIES YOU
world scenarios where the HAVE ACQUIRED AFTER THIS LESSON
assumptions of parametric
tests are violated? Understanding of non-parametric statistics
principles and concepts, Interpretation of results
from non-parametric tests and drawing meaningful
conclusions, Enhanced critical thinking and problem
solving skills in statistical analysis

PERSONAL REFLECTION

This lesson on non-parametric statistics provides

valuable skills to analyse data that doesn’t meet
parametric assumptions. It gives insights into non-
parametric tests like Wilcoxon rank-sum, Kruskal-
Wallis, and Kolmogorov-Smirnov, for comparing
groups, testing medians, and assessing goodness-
of-fit.

© Copyright 2021 MIVA Open University All Rights Reserved Key references:

Kolassa, J. E. (2017). Nonparametric Statistics: A Step-by-Step Approach. Wiley

Gibbons, J. D., Chakraborti, S. (2011). Nonparametric Statistical Inference (5th ed.). CRC Press