CONTENT

Abstract

Introduction

Objectives

Methodology

Subject matter

Parametric test

Non- parametric test

Characteristics of sum of rank test

How can we use to sum of rank test in educational research

Types of sum of rank test

Steps in educational research

Use in the sum of rank test

Advantages of sum of rank test

Disadvantages of sum of rank test

Conclusion

References

1
 Abstract
The Sum of Ranks Test, A Non-Parametric Alternative for Comparing Two Independent
Samples. The Sum of Ranks Test, also known as the Wilcoxon Rank-Sum Test or Mann-
Whitney U Test, is a non-parametric statistical test used to compare two independent
samples. This test is an alternative to the t-test and ANOVA, which assume normality and
equal variances. The Sum of Ranks Test is particularly useful when the data is ordinal,
skewed, or contains outliers.

The test involves combining the data from both samples, ranking the data from smallest to
largest, and calculating the sum of ranks for each sample. The test statistic is then calculated
as the smaller of the two sums of ranks. The test statistic is compared to a critical value or a
p-value is calculated to determine the significance of the results.

Keywords: Sum of Rank test, parametric test, Non- parametric test, Mann- Whitney U test,
Independent sample.

2
1.00 Introduction:
Statistics is the branch of mathematics that involves collecting, analyzing, interpreting,
presenting, and organizing data. It provides tools for making inferences or predictions about a
population based on a sample of data. Statistics is widely used in various fields such as
economics, business, health, and social sciences. Concepts in statistics include:

1. Descriptive Statistics: Summarizes and describes the features of a dataset using measures
like mean, median, mode, variance, and standard deviation.

2. Inferential Statistics: Uses sample data to make generalizations or predictions about a

population. This includes hypothesis testing, confidence intervals, and regression analysis.

3. Probability: A key part of inferential statistics, focusing on the likelihood of different

outcomes. It forms the foundation for statistical models and analysis.

4. Data Visualization: The graphical representation of data through charts, graphs, and plots
to help understand patterns, trends, and distributions.

5. Sampling: The process of selecting a representative subset of individuals or items from a

larger population to make inferences about the whole group.

6. Hypothesis Testing: A method used to determine if there is enough evidence in a sample

of data to support a particular belief or hypothesis about the population.

3
2.00 Objectives

The objectives of this seminar paper are-

1. To understand the concept of the sum of rank test and its type.
2. To know about how to use educational research.

4
3.00 Methodology:

This seminar paper is based on the descriptive method. Tools for data collected by various
secondary sources such as books and also internet sources.

5
4.00 Subject Matter:
Statistical tests can be divided into two types-
1. Parametric test
2. Non-parametric test

1. Parametric test:
Parametric test and Non-parametric test are two broad categories of statical hypothesis tests
used to make inferences about population based on sample data.
Parametric tests are based on assumptions about the distribution of the population from which
the sample is taken. A statistical test in which specific assumptions are made about the
population parameter. It is known as the parametric test. This test assumes that the data
follows a specific distribution, usually the normal distribution. These tests require certain
parameters to be estimated from the data such as mean, standard deviation, and variance.
Example of parametric test one ---- T-Test, Analysis of Variance, (ANOVA), regression
analysis, and chi-square test. Types of parametric test-

a) Student T- Test
b) Z- Test
c) ANOVA
d) Person co-relation coefficient

2. Non- Parametric Test:

Non-parametric test is also an important method of statistical analysis. The test that does not
require distribution to not the required assumption to be analysed. There is no requirement for
any distribution of population in the parametric test. Types of Non-parametric Test-
a) Maan Whitney U Test
b) Median test
c) Wilcoxon sign Rank test
d) Spearmen's Rank Difference Test.

Characteristics of sum of rank test-

The sum of ranks test, commonly known as the Wilcoxon rank-sum test or the Mann-
Whitney U test, is a non-parametric statistical test used to compare two independent samples.
Here are some key characteristics of this test:

1. Non-parametric: It does not assume that the data follows a specific distribution (e.g.,
normal distribution). This makes it useful when the assumptions for parametric tests are not
met.

6
2. Compares two independent samples: The test is typically used to compare two groups or
samples to determine if they come from the same population or if one group tends to have
larger values than the other.

3. Ranks the combined data: The data from both groups are combined, ranked, and then the
sum of the ranks for each group is calculated. This helps to assess the difference in
distributions between the two groups.

4. Tests for differences in central tendencies: The test evaluates whether one sample tends to
have higher or lower values than the other, typically checking for differences in medians or
overall distributions.

5. Hypotheses:

Null hypothesis (H₀): The two groups have the same distribution or median.

Alternative hypothesis (H₁): The two groups have different distributions or medians.

6. No assumptions about the underlying distribution: Since it's a rank-based test, there are
no assumptions about the shape of the distribution, making it suitable for ordinal data or
when the data are skewed.

7. Sensitivity to differences: The test is sensitive to differences in the overall shape of the
distributions, but it specifically evaluates the central tendency (e.g., median), not the specific
values in the distributions.

8. Uses ranks instead of raw data: The test involves converting raw data into ranks, which
reduces the influence of outliers and provides a more robust test when compared to mean-
based methods like t-tests.

In summary, the sum of ranks test is useful for comparing two independent samples when the
data is ordinal, non-normal, or when you cannot assume equal variances, making it a versatile
alternative to parametric tests.

How can we used to sum of Rank test in educational research:

The sum of ranks test, often referred to as the Mann-Whitney U test or the Wilcoxon rank-
sum test, is a non-parametric test used to determine whether there are differences between
two independent groups on a continuous or ordinal outcome. In educational research, this test
can be used in several ways:

7
i). Comparing Student Performance: To compare the scores of two different groups of
students, such as students in a traditional classroom versus an online learning environment.

If you want to see whether there is a significant difference in test scores between two
different teaching methods (e.g., lecture-based teaching versus project-based learning).

ii). Assessing the Effectiveness of Educational Programs: It can be used to assess whether
students from different educational programs (e.g., vocational vs. academic) perform
differently on standardized tests or final assessments. For example, researchers could assess
whether students who attended a summer enrichment program outperformed those who did
not.

iii). Comparing Group Differences Based on Demographic Factors: In cases where

students from different demographic groups (e.g., gender, socioeconomic status) are
compared in terms of their academic performance or engagement levels.

iv). Evaluating Intervention Impact: Used to compare test scores before and after an
intervention in different groups of students, where the intervention could be new teaching
materials or a specific learning activity.

v). Measuring Attitudes or Preferences: Educational research often requires assessing

attitudes or preferences, which can be ordinal data (e.g., strongly agree, agree, neutral,
disagree, strongly disagree). The sum of ranks test can compare how two groups rank their
attitudes toward a specific educational tool or strategy.

Steps in Educational Research using the Sum of Ranks Test:

1. Formulate Hypotheses:

Null Hypothesis (H0): There is no difference between the two groups.

Alternative Hypothesis (H1): There is a significant difference between the two groups.

2. Rank the Data: Combine the data from both groups and assign ranks to all the values,
starting with the lowest score.

3. Calculate the Sum of Ranks: Calculate the sum of the ranks for each group.

4. Compute the U statistic: The U statistic is calculated from the sum of ranks to determine
the likelihood of the observed difference occurring by chance.

5. Interpret the Results: Compare the U statistic to critical values from the Mann-Whitney
U distribution table to determine statistical significance.

8
This test is especially useful in educational research where the data may not be normally
distributed, or the sample size is small, making parametric tests (like t-tests) inappropriate.

Types of Sum of rank test:

The sum of ranks test is commonly used in non-parametric statistics to compare two or more
groups. There are several variations, depending on the context and the number of groups
involved. Here are some common types:

1. Mann-Whitney U test (Wilcoxon rank-sum test): This test is used to compare the ranks of
two independent samples. It tests whether there is a significant difference between the
distributions of the two groups. It is commonly used when the assumption of normality is not
met for the data.

2. Wilcoxon Signed-Rank Test: This test is used for comparing two related or paired
samples. It assesses whether there is a significant difference between the paired values. The
ranks are calculated for the differences between paired observations, and the test determines
if the sum of these ranks significantly deviates from zero.

3. Kruskal-Wallis H test: This test is an extension of the Mann-Whitney U test for comparing
more than two independent groups. It tests whether there is a significant difference between
the distributions of the groups. It ranks all the observations across the groups and checks
whether the ranks are equally distributed among the groups.

4. Friedman Test: The Friedman test is the non-parametric equivalent of the repeated
measures ANOVA.

Advantages of sum of rank test:

1. No Assumptions about Distribution: Unlike parametric tests, the rank-sum test does not
require the data to follow a specific distribution (such as a normal distribution). It is useful
when the assumptions of normality are not met.

2. Works with Ordinal Data: It can be used for ordinal data (data that has a meaningful order
but no consistent interval between values), making it more flexible than tests that require
interval or ratio data.

3. Robust to Outliers: Because the test uses ranks rather than the actual data values, it is less
affected by outliers or extreme values in the data.

9
4. Useful for Small Sample Sizes: The test is particularly effective with small sample sizes
where the central limit theorem may not apply and where parametric tests might not be
reliable.

5. Simple to Compute: The test is relatively simple to calculate, involving ranking the
combined data and comparing the sums of ranks between the two groups.

6. Non-parametric Nature: As a non-parametric test, it is suitable for data that does not meet
the assumptions of parametric tests, such as homogeneity of variance or normality.

7. Flexibility: It can be used in various scenarios, including comparing independent samples,

testing for differences in distributions, and even in cases where data is not continuous.

8. Test for Location Differences: The test is useful for determining whether there is a
significant difference in the location (median) of two distributions, especially when dealing
with non-normal data.

Disadvantages of Sum of rank test:

1. Less Power with Small Samples: When the sample size is small, the test might not have
enough power to detect significant differences between the groups, especially if the data is
sparse.

2. Doesn't Provide Exact Estimates: The test provides a p-value to indicate whether the
differences between groups are significant but does not give an estimate of the size or
magnitude of the difference between the two groups.

3. Assumes Similar Distribution: The test assumes that the distributions of both groups are
similar in shape, even though the underlying distributions might differ in other aspects, such
as scale.

4. Not Robust to Ties: When there are tied values in the data, the test can become less
reliable unless appropriate adjustments are made (e.g., using a correction for ties).

5. Only Compares Central Tendency: While it compares the ranks of data, it focuses on the
central tendency (median) of the groups rather than their full distributions, so subtle
differences in shape or variance might go undetected.

6. Requires Independent Samples: The test assumes that the two groups being compared are
independent of each other. If this assumption is violated, the results may not be valid.

10
7. Can Be Affected by Outliers: Although less sensitive than parametric tests, the sum of
ranks test can still be affected by extreme values in small datasets, leading to skewed results.

8. Limited to Two Groups: The test is generally designed for comparing only two
independent groups, and it doesn't extend easily to comparing more than two groups (without
modifications).

11
Conclusion

The Sum of Ranks Test, also known as the Wilcoxon Rank-Sum Test or Mann-Whitney U
Test, is a non-parametric statistical test used to compare two independent samples. This test is
an alternative to parametric tests, such as the t-test and ANOVA, which assume normality
and equal variances.

In conclusion, the Sum of Ranks Test is a useful non-parametric test for comparing two
independent samples. While it has some limitations, its advantages make it a popular choice
in many fields, including medicine, social sciences, and business. By understanding the
strengths and limitations of the Sum of Ranks Test, researchers, and practitioners can make
informed decisions about when to use this test and how to interpret the results.

12
REFERENCES:

1. Sharma N., Kulshreshtha R. (2019). “NTA-UGC-NET/JRF/SET PAPER 2

EDUCATION”. New Delhi, Arihant Publication (India) Limited.

2. Best W. J., Kahn V. J., Jha K. A. “Research in Education”. Pearson

13