SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC.

National Highway, Crossing Rubber, Tupi, South Cotabato

COLLEGE OF EDUCATION DEPARTMENT

___________________________________________________

LEARNING MODULE
FOR
M113: ADVANCED STATISTICS
_____________________________________________________

WEEK 13

M113: Advanced Statistics

SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 1 of 21
 COURSE OUTLINE
COURSE CODE : M113
TITLE : Advanced Statistics
TARGET POPULATION : All BSEd Mathematics Students
INSTRUCTOR : MS. CYRIL MEY S. QUICOY, LPT
[email protected] (09100424948)

Overview:

Introduction which includes a brief history of statistics, the uses of statistics and its categories,
deals on data collection and basic sampling techniques, presenting statistical data using frequency
tables and graphs, topics on the measures of central tendency and relative position while deals on
measures of variability, dispersion statistics, analysis of variance, demographic methods and vital
statistics and chi-squared test.

Objectives:

Aims to develop in the students the ability to think in a logical, critical, and systematic manner.

The following are the topics to be discussed

Week 13 RELATIVE MEASURE OF VARIATION

Week 14 PERCENTILE RANK
Week 15 CORRELATION
Week 16 COMPUTATION OF CORRELATION
Week 17 SPEARMAN RHO COEFFICIENT

Instruction to the Learners

Each chapter in this module contains a major lesson involving the statistical concepts. The units
are characterized by continuity, and are arranged in such a manner that the present unit is related to
the next unit. For this reason, you are advised to read this module. After each unit, there are exercises
to be given. Submission of task given will be every Monday during your scheduled class hour.

M113: Advanced Statistics

SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 2 of 21
 WEEK 13
NON-PARAMETRIC STATISTICS
A. Basic Concepts
Two techniques of inferring in modern statistics are parametric and non-parametric
statistics. Parametric statistics is based on the assumption that the data are drawn from a
normally distributed population. It may also be based on the assumption that the two sets of
data may have the same mean, median, or variance.
Non-parametric statistics, however, does not require many assumptions, such as
normality of data. These distribution-free techniques yield conclusions regardless of the shape
of the population distributions. Table 1 compares some parametric and non-parametric statistical
tests under different cases.
Table 11.1. Comparison of Parametric and Non-Parametric Tests
Cases Parametric Tests Non-parametric Tests
One Sample One-sample t-test or z-test Sign test
Two independent samples Two-sample independent t-test Wilcoxon-Mann-Whitney rank-
or z-test sum test
Two dependent samples Two paired t-test or z-test Wilcoxon signed-rank test sign
test
More than two groups (one One-way ANOVA Krsukal-Wallis one-way ANOVA
factor) by ranks
More than two groups (two Two-way ANOVA Friedman’s two-way ANOVA by
factors) ranks
Correlation Pearson r Spearman rank-order
correlation

If parametric and non-parametric tests are both applicable to the data, the researcher should
use the parametric test when normality is justifiable. On the other hand, if normality cannot be
established, the non-parametric test should be used.

B. THE SIGN TEST

The sign test is appropriate when the direction of the differences between two measures,
rather than the quantitative measures, are considered as data. The differences can be
represented by the symbols plus (+) and minus (-) signs. All tied cases are not included in the
analysis. This test is preferable for small samples of size n˂30.
The null hypothesis Ho can be stated in terms of the sings of the differences X i − y i. If
the data value is above the given specified measure, it is assigned a plus sign, and, if below, a
minus sign. If the numbers of plus and minus signs are equal, then the null hypothesis is true.
The alternative hypothesis, H 1 can be stated as one-tailed or two-tailed. It commonly
states that one sign (+ or -) will occur more frequently or that the frequencies in which the two
signs occur is significantly different.

Four steps are used in the sign test:

Step 1: Account for the difference between the two members of each pair. Assign (-) minus for
M113: Advanced Statistics
SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 3 of 21
 the decrease and (+) plus for the increase.
Step 2. Count the numbers of pluses and minuses and determine the test value S of n which is
equal to the smaller number of plusses and minuses. Disregard the tied cases, that is, those
with no differences.
Step 3. For small and large samples, the following should be considered:
a. For a one-tailed and two-tailed test; if n ≤ 25, use Table A.6 for Critical Values Sa
which are obtained from the binomial distribution, where n is the total number of +
and – signs. The test value S, is the smaller number of plus or minus signs. If S ≤ Sa ,
the null hypothesis is rejected.
b. If n ˂ 25, use the z-statistic as an approximation to the binomial distribution using the
following computations:

The correction for continuity is where x + 0.5 is used

n n
when x ˂ and x – 0.5 when x > .
2 2
Step 4. If x is in the critical region, the Ho is rejected.

NOTE: The probability can also be computed using the following formula for binomial
distribution:

Where n is the number of paired observations which shows a difference and y= n-x. if the
probability is less than or equal to a, Ho is rejected.

Example:
Some graduates attended an intensive review class for the Licensure Examinations for
Teachers (LET). They took a pre-test and a similar post-test at the start and at the end of the review,
respectively. The following table shows the scores of seven reviewees in the pre- and post- tests.
Determine whether the intensive review class improves the scores of the seven reviewees at a= 0.05.
Mar Carla Jean Carlo John Michael Butch Vic May Cel Ja Tom
y
Pre-test, Xi 84 74 70 75 83 82 75 83 82 73 76 80
Post-test, Y i 83 79 85 88 85 88 80 87 84 79 81 83

SOLUTION: Use the steps in hypothesis testing.

Step 1. State the null and the alternative hypotheses
Ho: The intensive review class has no effect on the scores in the post test. That is, there
M113: Advanced Statistics
SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 4 of 21
 is no significant difference in the numbers of plus and minus signs or the frequencies of
pluses and minuses are the same.
H1: The intensive review class increased the scores of the reviewees in the post-test.
That is, there are more plus signs than minus signs or pluses occur more frequently than
minuses.
Step 2. Set the significance level.
Let a = 0.05 and n be the number of paired observations which showed a difference in
their scores in the pre-test and post-test. Since there are no tied observations, n= 7.
Step 3. Choose an appropriate statistical tool or test statistic.
The sign test is chosen for this study. A plus sign denotes a success of the event
and a minus sign denotes a failure of the event. The answer to the question is
dichotomous, that is, “increase” or “ does not increase”; thus, the associated probability
of occurrence of values as large as x (+ or -) is given by the binomial distribution for

1
p=q= .
2
Step 4. Establish critical and non-critical regions.
Since the alternative hypothesis predicts the direction of the difference, use the
one-sided to the right test. The rejection region is S≤ Sa with a = 0.05 and n=12 from

table A.6. Sa = 2. Thus, the critical region is S≤ 2.

Step 5. Compute the test statistic.
Mar Carla Jean Carlo John Michael Butch Vic May Cel Ja Tom
y
Pre-test, Xi 84 74 70 75 83 82 75 83 82 73 76 80
Post-test, Y i 83 79 85 88 85 88 80 87 84 79 81 83
Sign of
(post-test - + + + + + + + + + + +
pre-test) [

y i−x i ¿

The test statistic is S= 1, which is the number of minus signs.

Step 6. Decide whether to reject or fail to reject the null hypothesis.
The rejection region is S≤2. Since S=1 is greater than Sa = 2, the null hypothesis

is rejected in favor of the alternative hypothesis, H 1.

Step 7. Conclusion
The LET intensive review class had a significant effect on the post-test scores of
the seven graduates.

M113: Advanced Statistics

SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 5 of 21
 ACTIVITY #13
Direction: Read and understand this module. Provide what is being asked. Write your answer in a long
bond paper with proper template, take a photo of your written answer and send it to our M113 Google
Classroom. Template of your task is as follows:

NAME: QUICOY, CYRIL MEY S. SUBJECT CODE: M113

COURSE & SET: BSED-MATH 3 DATE: September 1, 2021

NOTE: See to it that your work is readable.

1. The running times (in minutes) of 15 participants in a 3-km fun run for the rehabilitation
of the Taguibo Watershed are the following: 34, 15, 23, 32, 29, 40, 21, 20, 36, 37, 17,
25, 30, 32, and 33. Apply the sign test at a=0.05 to determine whether the mean running
times differ from the expected running time of 20 minutes for this run.

End of thirteenth week

---------------------------------------------Nothing Follows--------------------------------------

M113: Advanced Statistics

SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 6 of 21
 SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC.
National Highway, Crossing Rubber, Tupi, South Cotabato

COLLEGE OF EDUCATION DEPARTMENT

___________________________________________________

LEARNING MODULE
FOR
M113: ADVANCED STATISTICS
_____________________________________________________

WEEK 14
M113: Advanced Statistics
SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 7 of 21
 WEEK 14
WILCOXON SIGNED-RANK TEST
The sign test uses only the plus and minus signs of the differences but does not consider the
magnitude of the differences. The Wilcoxon signed-rank test is a test which considers not only the signs
but also their magnitudes.

The steps used in the Wilcoxon signed-rank test are as follows:

Step 1. Obtain the differences as in the sign test.
Step 2. Take the absolute value of the differences and arrange them from lowest to highest.
Step 3. Assign the number 1 to the smallest value, the number 2 to the next smallest, and so on up to
n, where n is the number of non-zero differences. In case two or more values are equal, obtain the
mean of the numbers assigned to the values and this will be used as the ranks of the equal values. For
example, if the second and the third values are equal, then their ranks will be the mean of 2 and 3
which is 2.5.
Step 4. Assign the + or – signs to the ranks according to the signs of the differences.
+¿¿ −¿ ¿
Step 5. Obtain the sum of all the positive ranks ( S and the sum of the negative ranks ( S .
Step 6. Get the absolute values of the two sums, that is,│ S+¿│ ¿ and │ S−¿ │ ¿. The test statistic will be the
smaller value, denoted by S.

For n ≤ 30, use Table A.7 for the critical values of the Wilcoxon signed-rank test. The null hypothesis is
rejected if the test statistic is less than or equal to the critical values.

For n ≤ 30, use the z-statistic in the following computations:

For tied ranks, apply the following formula:

Where g is the number of groupings of difference and s j is the number of tied ranks or groupings.
The null hypothesis can be stated as:
H 0 :μ 1=μ2
The alternative hypothesis can be written as:

M113: Advanced Statistics

SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 8 of 21
 +¿¿ −¿ ¿
Ho will be rejected when S is different from S at a given significance level.

Example:
The weights (in kilograms) of cabbages harvested from two different plots labeled Plot 1 and
Plot 2 were compared. Organic fertilizer was applied on Plot 1 and chemical fertilizer was used on Plot
2. The data on 8 cabbages per plot are given in the following table. Use the Wilcoxon signed-rank test
to test the hypothesis that organic fertilizer gives a better yield than chemical fertilizer at 0.05 level of
significance.
Plot 1 (Organic Fertilizer) 0.63 0.61 0.68 0.66 0.62 0.65 0.71 0.67
Plot 2 (Chemical Organizer) 0.66 0.67 0.65 0.68 0.68 0.60 0.61 0.64

SOLUTION:
Step 1. State the null and the alternative hypotheses.
+¿¿ −¿ ¿
Ho: S >S (The mean weights of the cabbages grown in plot 1 are greater than the mean
weights of those grown in plot 2.)
H 1 : S+¿¿ = S−¿ ¿ (There is no significant difference in the mean weights of cabbages grown in plot
1 and those grown in plot 2.)
Step 2. Set the significance level.
It is given that a = 0.05 and n = 8.
Step 3. Choose an appropriate statistical tool or test statistic.
In this problem, the Wilcoxon signed rank test will be used.
Step 4. Establish critical and non-critical regions.
Using Table A.7, the critical region is S ≤ 6 for n = 8.
Step 5. Compute the test statistic.

Step 6. Decide whether to reject or fail to reject the null hypothesis.

Since S=17 and the critical region is S≤ 6, there is no sufficient evidence to reject Ho.
Step 7. Conclusion
Therefore, there is no significant difference in the mean weights of the cabbages grown with
organic and chemical fertilizers.

M113: Advanced Statistics

SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 9 of 21
 ACTIVITY #14
Direction: Read and understand this module. Provide what is being asked. Write your answer in a long
bond paper with proper template, take a photo of your written answer and send it to our M113 Google
Classroom. Template of your task is as follows:

NAME: QUICOY, CYRIL MEY S. SUBJECT CODE: M113

COURSE & SET: BSED-MATH 3 DATE: September 1, 2021

NOTE: See to it that your work is readable.

1. The heights of eight randomly chosen male and female preschoolers were measured. The data
(in centimeters) are shown in the following table. At a = 0.05, is there a significant difference in
the mean heights of the male and female preschoolers?
Male 82 81 77 72 71 63 67 79
Female 72 79 78 80 89 76 79 78

End of fourteenth week

---------------------------------------------Nothing Follows--------------------------------------

M113: Advanced Statistics

SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 10 of 21
 SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC.
National Highway, Crossing Rubber, Tupi, South Cotabato

COLLEGE OF EDUCATION DEPARTMENT

___________________________________________________

LEARNING MODULE
FOR
M113: ADVANCED STATISTICS
_____________________________________________________

M113: Advanced Statistics

SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 11 of 21
 WEEK 15

WEEK 15
WILCOXON-MANN-WHINEY RANK-SUM TEST
The Wilcoxon-Mann-Whitney rank-sum test may be used to determine if two
independent samples came from the same population. This test is the non-parametric
counterpart of the t-test.
Suppose that the two independent sample X and Y were drawn from population with the
same distribution, where the number of observations is X is less than or equal to the number of
observations in Y. Let θ x and θ y be the medians of the two samples X and Y, respectively. The
hypotheses may be stated as follows:
Null hypothesis:
Ho: θ x =θ y (The two independent samples X and Y are drawn from the same
population.)
Alternative Hypothesis:
The alternative hypothesis can be stated as follows:
H 1: θ x ≠θ y (two tailed)
H 1: θ x >θ y (left-tailed)
H 1: θ x ˂θ y (right-tailed)
To apply the Wilcoxon-Mann-Whitney rank-sum test, perform the following steps:
Step 1. Let m be the number of observations from sample; X and n be the number of
observations from sample Y, where m≤n.
Step 2. Put together the observations from the two samples; arrange them in ascending
order; and rank them from lowest to highest, that is, assign the number 1 to the smallest
value, number 2 to the next smallest, and so on up to m + n. In case two or more vales
are equal, obtain the means= of the numbers assigned to the values and use this as the
ranks of the equal values. Example, if the second and the third values are equal, then
their ranks will be the mean of 2 and 3 which is 2.5.
Step 3. Obtain the sum of the ranks W x and W y for the samples X and Y,, respectively.
That is:
m
W x = sum of ranks X with m observations = ∑ rank ( x i)
1

m
W y = sum of ranks X with n observations ¿ ∑ rank ( y i)
1

For small samples ( m≤10, n≤10), the null hypothesis will not be rejected after solving
W x and W y if the sums are approximately equal.

M113: Advanced Statistics

SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 12 of 21
 For larger samples ( m≤10, n≤10), the sampling distribution of W or the smallest rank
sum approaches a normal distribution with:

With the continuity correction factor +0.5 for left-tailed test and -0.5 for right tailed test. For the
two-tailed test,

Example:
Six students from an urban high school and seven students from a rural high school
were chosen as samples of a study on whether the performance in mathematics depended on
the demographic location of the school. They were given a test to assess their performance in
math. The test scores of the thirteen students are shown in Table 2. Perform the Wilcoxon-
Mann-Whitney rank-sum test to determine whether there is a significant difference in the
performance in math between the students from the urban high school and those from the rural
high school at 0.05 level of significance.

Table 11.2. Scores of Rural and Urban High School Students

Rural High School (Y) 88 89 91 95 96 99 100
Urban High School (X) 82 85 87 90 92 93 -

SOLUTION:
Step 1. State the null and the alternative hypotheses.
Ho: There is no significant difference in the performance in mathematics between
the students from the rural high school and these from the urban high schools.
H 1 : The students from the rural high school perform better than the students from
the urban high school.
Step 2. Set the significance level.
Let a = 0.05 be the level of significance.
Step 3. Choose an appropriate statistical tool or test statistic.
The Wilcoxon-Mann-Whitney rank-sum test is used with m= 6 and n=7.
M113: Advanced Statistics
SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 13 of 21
Step 4. Establish critical and non-critical regions.
The null hypothesis is rejected if W x≠ W y.
Step 5. Compute the test statistic.
The following table provides the ranks of the test scores:

Rural High School Urban High School

Scores Rank Scores Rank
88 4 82 1
89 5 85 2
91 7 87 3
95 10 90 6
96 11 92 8
99 12 93 9
100 13
W x =62 W y =29

From the preceding table, W x =62 andW y= 29. Thus, W x ≠ W y .

Step 6. Decide whether to reject or fail to reject the null hypothesis. Since W x and W y, are not
equal, Ho is rejected.
Step 7. Conclusion
Thus, the students from the rural high school performed better than the students from
the urban high school.

M113: Advanced Statistics

SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 14 of 21
 ACTIVITY #15
Direction: Read and understand this module. Provide what is being asked. Write your answer in a long
bond paper with proper template, take a photo of your written answer and send it to our M113 Google
Classroom. Template of your task is as follows:

NAME: QUICOY, CYRIL MEY S. SUBJECT CODE: M113

COURSE & SET: BSED-MATH 3 DATE: September 1, 2021

NOTE: See to it that your work is readable.

1. The manager in a call center gave an operation manual to two groups of employees: those from
the day shift and those from the night shift. Each group was then tested on operation
procedures. The scores are shown in the following table. The manager felt that employees in
the day shift performed better. Compare the mean test scores of the two groups of employees
as tabulated and report your conclusion. State the hypothesis. Set a= 0.05.
Employee Test Scores
Day-Shift 77 87 72 87 82 80 71 79 80 78 77 79 83
Night-Shift 82 79 70 63 74 83 76 78 81 72 71 74 62 64

End of fifteenth week

---------------------------------------------Nothing Follows--------------------------------------

M113: Advanced Statistics

SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 15 of 21
 SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC.
National Highway, Crossing Rubber, Tupi, South Cotabato

COLLEGE OF EDUCATION DEPARTMENT

___________________________________________________

LEARNING MODULE
FOR
M113: ADVANCED STATISTICS
_____________________________________________________

WEEK 16

WEEK 16
M113: Advanced Statistics
SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 16 of 21
 MINI-RESEARCH PROJECT

Report Format for the Mini Research

Executive Summary

1. Introduction

1.1 Introduction

1.2 Industry Analysis (if you are addressing particular industry)

1.3 Research Problem

1.4 Research Objectives

1.5 Limitations of the study

2. Literature Review

2.1 Theories, concepts & models

2.2 Previous Research

3. Research Methodology

3.1 Introduction

3.2 Theoretical Framework (Model with dependent and independent variables)

3.3 Research Design

3.4 Hypothesis (IF ANY)

3.5 Data Collection Methods

3.6 Sampling Method (including sample size)

4. Data Presentation & Analysis

4.1 Background Information Analysis

4.2 Hypotheses Testing

5. Conclusion

Appendices

References

End of sixteenth week

---------------------------------------------Nothing Follows--------------------------------------

M113: Advanced Statistics

SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 17 of 21
 SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC.
National Highway, Crossing Rubber, Tupi, South Cotabato

COLLEGE OF EDUCATION DEPARTMENT

___________________________________________________

LEARNING MODULE
FOR
M113: ADVANCED STATISTICS
_____________________________________________________

WEEK 17

WEEK 17

M113: Advanced Statistics

SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 18 of 21
 PRESENTATION OF MINI-RESEARCH

M113: Advanced Statistics

SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 19 of 21
M113: Advanced Statistics
SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 20 of 21
 End of seventeenth week
---------------------------------------------Nothing Follows--------------------------------------

M113: Advanced Statistics

SOUTH EAST ASIAN INSTITUTE OF TECHNOLOGY, INC. Page 21 of 21