MATHEMATICS IN THE MODERN WORLD

SECTION 2.2.1 APPLICATIONS OF DESCRIPTIVE STATISTICS

DESCRIPTIVE STATISTICS Descriptive Statistics Inferential Statistics
▪ Concerned with ▪ Concerned with making
Statistical tools derived from
describing the target inferences from the
mathematics are useful in processing and population sample and generalize
▪ Organize, analyze, and them to the population
managing data. Selection of appropriate tools present the data in a ▪ Compare, test
and efficient use of these tools can help people meaningful manner hypothesis, and predict
▪ Results are shown in future outcomes
organize, analyze, and interpret data. One of the form of charts, tables, ▪ Results are shown as
useful statistical tools in data management is and graphs statistics and probability
▪ Tools: measures of values
descriptive statistics. Descriptive statistics central tendency, ▪ Tools: hypothesis tests,
variation, and position analysis of variance
deals with organization, presentation, and
analysis of data that help describe, show, or Table 2.1 Descriptive Statistics vs Inferential Statistics
summarize data in a meaningful way. It covers Table 1 summarizes the differences
measures of central tendency, measures of between descriptive statistics and inferential
variation, measures of position, and graphical statistics in terms of their nature, purpose, data
presentation of data. Although these concepts presentation, and data analysis tools.
appear to be the same as the statistics taught at Which of the following situations below
Junior High School mathematics, the intention will require the use of descriptive statistics?
of this topic is to highlight students’ skills in Situation 1: A sports analyst wants to know the
selecting appropriate descriptive statistics and performance of basketball players based on the
interpreting statistical results. Furthermore, tallied numbers such as points scored, blocks,
and rebounds during a game series.
this topic is focused on the applications of
Situation 2: A health expert plans to investigate
descriptive statistics in real-life situations.
what public measures would prevent a disease
from spreading and contribute to its eradication.

Situation 3: A teacher wishes to know the

achievement of students in the Mathematics in
the Modern World after the midterm period.

If you answered situations 1 and 3, you are

correct!
Source: https://www.freepik.com/free-vector/statistics-bar-doughnut-line-
charts-template_3034283.htm#page=1&query=statistics&position=2
HOLY ANGEL UNIVERSITY I SCHOOL OF ARTS AND SCIENCES 1
 MATHEMATICS IN THE MODERN WORLD
DATA PRESENTATION
One of the important processes involved in
data management is presentation of data. In
descriptive statistics, data are summarized,
organized, and communicated using a variety of
visual tools such as tables, diagrams, charts, and
statistical graphs. Common statistical tools
include line frequency distribution, line graph,
bar graph, pie graph, and scatter plots. The use of
these graphs depends on the purpose of the and
nature of the data. This topic presents ways on Source: https://www.sws.org.ph/swsmain/artcldisppage/?artcsyscode=ART-
20160223095529
how to organize and present data in charts and
graphs. Bar Graph
▪ presents the data by using horizontal or
vertical bars whose heights or lengths are
determined by the frequencies of the data
▪ can be used to compare data for two or
more groups

Source: https://cyberbabble.blog/2019/04/22/the-importance-of-data-
presentation/

Here are some common statistical

graphs used in data presentation with
illustrative examples.
Time Series Graph
▪ represents data that occur over a specific
period; to look for pattern or trend that
occurs over the period of time Source: https://www.geekwire.com/2012/chart-pays-seattle-software-
engineers/
▪ two or more data sets can be compared on
the same time series graph

HOLY ANGEL UNIVERSITY I SCHOOL OF ARTS AND SCIENCES 2

 MATHEMATICS IN THE MODERN WORLD
Pie Graph
Frequency Distribution
▪ is a circular graph which is divided into
▪ is the organization of data in tabular form
sectors or wedges according to the
using classes and frequencies
percentage of the frequencies in each
category of the distribution of a nominal or
categorical variable
▪ used in showing relationship of the parts to
the whole by visually comparing the sizes of
the sections
Common Causes of Catastrophes in Business

Source: Estrada, L. (2015). Effecting change effecting change on students’

performance and attitude toward problem solving in mathematics through
schema-based instruction. Unpublished thesis.

You can make formal, creative, and

presentable graphs and charts using computer
software or applications. Canva and MS Excel are
Source: https://oncall258.com/business-continuity-planning-alexandria-va/
the most used tools in creating charts or graphs

Scatter Plot for data presentation. You need to choose the

▪ is a graph of ordered pairs of data values most appropriate statistical graph to

that is used to describe the relationship that communicate the results or information.

exists between two variables Based on what you have learned, identify
the most appropriate statistical graph to be used
in presenting the following data and state your
reason.
a. Common cyber crimes
b. Monthly maximum temperature
recorded in 2019
c. Population of ASEAN countries in 2019
d. Earth’s composition of gases
e. Number of HAU Passers in Mechanical
Engineer Examination from 2015-2019
Source: https://www.mymarketresearchmethods.com/types-of-charts-
choose/
HOLY ANGEL UNIVERSITY I SCHOOL OF ARTS AND SCIENCES 3
 MATHEMATICS IN THE MODERN WORLD

DESCRIPTIVE MEASURES Weighted Mean

The previous topic shows how you can ▪ is determined when the values are not
organize, present, and communicate equally represented (e.g. general weighted
information using various statistical graphs. grade
This topic will show you the statistical tools ▪ is obtained by multiplying each value by its
that can be used to summarize quantitative corresponding weight, divided the sum of the
data. The most familiar of these tools are products by the sum of the weights
measures of central tendency and measures of
Mode
variation. These measures may be considered
parameter or statistic. ▪ is the value that occurs most often in the data
set
PARAMETER vs STATISTIC
Parameter is a Statistic is a ▪ is denoted by 𝑥̂
characteristic or characteristic or ▪ is obtained by identifying which value has the
measure obtained all measure obtained from
greatest frequency
the values from a sample of a population.
population ▪ a data set may be unimodal (has 1 mode),
bimodal (has 2 modes), multimodal (has 3 or
Measures of Central Tendency more modes) or no mode
Measures of central tendency are also Median
called measures of average. They represent the
▪ is the middle most value or midpoint of a data
center of distribution or the most typical case.
set
These measures include mean, median, and
▪ is represented by 𝑥̃
mode. Do you still remember them?
▪ is obtained by arranging the values in
Mean ascending or descending order and selecting
▪ a.k.a. arithmetic average the middle value if the number of values is
▪ is denoted by 𝑥̅ (sample mean) or 𝜇 odd or get the mean of the two middle values
(population mean) if the number of values is even
▪ is obtained by getting the sum of the values
In APA format, the symbols M and Mdn are used to report the
in a data set, divided by the number of mean and median of data, respectively. There is no symbol for
values mode.

▪
HOLY ANGEL UNIVERSITY I SCHOOL OF ARTS AND SCIENCES 4
 MATHEMATICS IN THE MODERN WORLD

Measures of Variation Variance Standard Deviation

The measures of variation or dispersion
Population ∑(𝑥𝑖 − 𝜇)2 ∑(𝑥𝑖 − 𝜇)2
𝜎2 = 𝜎=√
tell us how the data values are dispersed or 𝑁 𝑁

scattered. They indicate the degree to which ∑(𝑥𝑖 − 𝑥̅ )2

𝑠2 = ∑(𝑥𝑖 − 𝑥̅ )2
Sample 𝑛−1 𝑠=√
quantitative data tend to spread about an 𝑛−1

average value. Three common measures of Legends: 𝜇 – population mean; N – population size; 𝑥̅ – sample mean; n
– sample size; xi – individual value
variation are range, variance, and standard
Table 2.2 Formulas for Variance and Standard Deviation
deviation.

Range
Example
▪ is obtained by computing the difference of 1. A sample of 8 drivers shows the time (in
the highest value and the lowest value in a minutes) that they spent in rush-hour traffic
data set that day. Given the data set below, compute
▪ is denoted by R for the mode, median, mean, range, standard
deviation, and variance.
Variance
Driver A B C D E F G H
▪ is obtained by getting the average of the Time 52 61 53 60 55 47 60 44
squares of the distance each value is from
Solutions:
the mean
a. Mode = 60 minutes
▪ is represented by 𝜎 2 (population variance)
60 has the highest frequency.
or s2 (sample variance)
b. Median = 54 minute
▪ is the square of standard deviation
Arrange the values in ascending order: 44,
Standard Deviation 47, 52, 53, 55, 60, 60, 61
Since the number of values is (8) even, the
▪ is obtained by the square root of variance
average of the two middle values (53 and
▪ is represented by 𝜎 (population standard
55) is 54.
deviation or s (sample standard deviation)

The formulas for standard deviation are

presented in Table 2. Take note that variance is
the square of standard deviation.

HOLY ANGEL UNIVERSITY I SCHOOL OF ARTS AND SCIENCES 5

 MATHEMATICS IN THE MODERN WORLD

c. Mean = 54 minutes Data Analysis in MS Excel

44 + 47 + 52 + 53 + 55 + 60 + 60 + 61 For larger population size or sample size, you
𝑥̅ =
8
can explore MS Excel in computing for the mean,
432
𝑥̅ = = 54
8 standard deviation, and variance.
Watch the video: How to Find Average and Standard
Deviation in Microsoft Excel 2017
d. Range = 17 minutes
https://www.youtube.com/watch?v=BGq8kuffR_Q
highest value – lowest value = 61 – 44 = 17
Practice makes perfect. Try to answer the
e. Variance = 39.4 minutes following problems.
∑(𝑥𝑖 −𝑥̅ )2 1. Listed below are the number of calories in
Use the formula 𝑠 2 = 𝑛−1
selected salads at several restaurants. Find
𝑥𝑖 (𝑥𝑖 − 𝑥̅ )2
the mean, median, and mode for the data.
44 (44 – 54)2 = 100
47 (47 – 54)2 = 49 1165 1180 1090 900 1270
52 (52 – 54)2 = 4 985 855 1120 900 1180
53 (43 – 54)2 = 1 900 1250 855 730 1595
55 (55 – 54)2 = 1 2. The grades of 10 students after completing an
60 (60 – 54)2 = 36 online course are presented below.
60 (60 – 54)2 = 36 Determine the range, variance, and standard
61 (41 – 54)2 = 49
deviation for the data.
n=8 𝛴(𝑥𝑖 − 𝑥̅ )2 = 276
1.00 1.50 1.75 1.50 1.25
1.50 2.25 2.25 1.00 2.00
2 276
𝑠 = 8−1 ≈ 39.4 square minutes 3. A testing laboratory wishes to test two
f. Standard deviation = 6.3 minutes experimental brands of paint to see how long

∑(𝑥𝑖 −𝑥̅ )2
each will last before fading. The testing lab
Use the formula 𝑠 = √ 𝑛−1 makes 6 trials for each brand. The results (in

276 months) are shown below. Which paint brand

𝑠 = √8−1 ≈ 6.3 minutes
is better? Support your answer.
Note that standard deviation is the square root Brand A Brand B
10 35
of variance.
60 45
s = √39.4 ≈ 6.3 50 30
30 35
40 40
20 25

4.
HOLY ANGEL UNIVERSITY I SCHOOL OF ARTS AND SCIENCES 6
 MATHEMATICS IN THE MODERN WORLD

Supplemental Readings

Descriptive vs. Inferential Statistics: What’s the Difference?

by Statology
https://www.statology.org/descriptive-inferential-statistics/

Descriptive statistics: Use these tools to analyze data vital to practice-improvement projects
by Corner, B. & Johnson, E.
Source: https://www.myamericannurse.com/wp-content/uploads/2017/11/ant11-Research-101-
1017a-1.pdf

Presenting Numerical Data

by the Student Learning Development, University of Leicester
Source: https://www2.le.ac.uk/offices/ld/all-resources/numerical-data/numerical-data

References
Singh, S. (2018). Descriptive and inferential Singh. https://towardsdatascience.com/statistics-
descriptive-and-inferential-63661eb13bb5

Blumann, A. (2016). Elementary statistics: A step by step approach, A brief Version (7 th ed.). McGraw-
Hill Education.

HOLY ANGEL UNIVERSITY I SCHOOL OF ARTS AND SCIENCES 7

 MATHEMATICS IN THE MODERN WORLD

SECTION 2.2.2 APPLICATIONS OF HYPOTHESIS TESTING

HYPOTHESIS TESTING
Statistics, as a mathematical tool, is
designed to help the researcher or statistician in
different fields to make decisions about the
outcomes of potentially important body of
knowledge for planning, development, and
evaluation. Inferential statistics is most
appropriate in situations that call for testing,
comparing, and predicting outcomes. One
crucial process in conducting inferential
Source:
statistics is hypothesis testing. In this topic, you
will explore the applications of hypothesis
testing. Statistics in Real World
When you were in Senior High School, Today many higher education institutions offer
online courses because of the impact of the
you already had an introductory course on Coronavirus disease 2019 (COVID-19) pandemic. With
parametric statistics through the mathematics the increased use of this learning platform, some
subject “Statistics and Probability.” In this topic questions about the effectiveness of online learning
have been raised. For instance,
your knowledge and skills about statistical
▪ Do students perform better in online learning
hypothesis testing will be further enhanced.
setting than in the face-to-face classroom
Hypothesis testing can be likened to a court setting?
trial. We assume that the defendant is either ▪ Do gender, types of interaction, and technology
innocent or guilty, and the defendant will be facility influence student satisfaction with online
learning?
convicted or acquitted.
▪ Is there is a significant relationship between self-
Your MMW teacher assumes that you are efficacy and online learning environments?
equipped with basic knowledge about statistical To answer these questions, we need to work with
hypothesis testing. Let us begin with defining data set and perform data analysis using various
statistical tests through hypothesis testing
some terms related to hypothesis testing.

HOLY ANGEL UNIVERSITY I SCHOOL OF ARTS AND SCIENCES 8

 MATHEMATICS IN THE MODERN WORLD

Definition of Terms Type II Error

Hypothesis ▪ occurs when a false null hypothesis is
▪ is an assumption, a claim, or a proposed accepted or not rejected, a.k.a. Beta error
explanation for an observation or One-Tailed Test
phenomenon
▪ is a test that indicates that the null
Null Hypothesis hypothesis should be rejected when the
▪ is a claim that indicates the absence of test statistic value is in the critical
relationship or difference between two region on one side of the mean, a.k.a.
values, denoted by symbol Ho directional test

Alternative Hypothesis Two-Tailed Test

▪ is a claim that indicates the presence of ▪ is a test that indicates the null hypothesis
relationship or difference between values, should be rejected when the test statistic
denoted by symbol Ha or H1 value is in either of the two critical
regions, a.k.a. non-directional test
Hypothesis Testing
Confidence Level
▪ is a decision-making process for
evaluating or testing claims about a ▪ is the probability that a parameter lies
population based on information obtained within the specified interval estimate of
from samples the parameter

Level of Significance Critical Value

▪ is the maximum probability of committing ▪ is a value that separates a critical region

type I error, denoted by the Greek (rejection region) from acceptance region
alphabet alpha (𝛼) in a hypothesis test, usually presented in
tables
Type I Error
Statistic
▪ occurs when a true null hypothesis is
rejected, a.k.a. Alpha error ▪ is a value obtained by using all data value
from a sample

HOLY ANGEL UNIVERSITY I SCHOOL OF ARTS AND SCIENCES 9

 MATHEMATICS IN THE MODERN WORLD

Steps in Hypothesis Testing For the sample size, according the Central
Table 2.3 summarizes the steps to be Limit Theorem, at least 30 samples is
undertaken when conducting statistical recommended. The test statistics to be employed
hypothesis testing. Every hypothesis-testing depends on the nature of data and the purpose of
context starts with declaration of hypothesis. test. What are the scales of measurement of the
There are two types of statistical hypotheses data? Are data continuous? Is it a test of
for each context: the null hypothesis and the association, comparison, or difference? In the
alternative hypothesis. next subtopic, common statistical tests will be
presented.
Steps in Statistical Hypothesis Testing
The numerical value obtained from a
1 State the null hypothesis and alternative
hypothesis statistical test (called the computed value or test
2 Set the level of significance and determine the
statistic) is compared with critical value to
direction of the tests
3 Collect data decide on the rejection or acceptance of null
4 Calculate a test statistic and determine the hypothesis. There are four possible outcomes as
critical value
5 Decide on the rejection or acceptance of null shown in Table 2.4 in hypothesis-testing
hypothesis situation.
6 Draw a conclusion
Finally, based on the decision, we have to
Table 2.3 Formulas for Variance and Standard Deviation
draw a conclusion. When drawing a conclusion
Moreover, in hypothesis testing, the or summary of results of a statistical study, it is
researcher decides what level of significance important to state whether there is enough
(𝛼) to use. In research 𝛼 is usually set at 0.01 or evidence to reject the null hypothesis or support
0.05. After a significance level is chose, the type the alternative hypothesis.
of test is determined: one-tailed test or two-
For you, which hypothesis-testing step is
tailed test. The type of test depends on the
difficult to accomplish? Why?
statement of alternative hypothesis. If the
alternative hypothesis involves the symbol > or True Ho False Ho
Type I Correct
<, then the test requires one-tailed test. If the Reject Ho
Error Decision
alternative hypothesis involves, involves the Accept Ho
Correct Type II
Decision Error
symbol ≠, then the test requires two-tailed test. Table 2.4 Four Possible Outcomes of a Hypothesis Test

HOLY ANGEL UNIVERSITY I SCHOOL OF ARTS AND SCIENCES 10

 MATHEMATICS IN THE MODERN WORLD

FURTHER READING 2. A medical researcher is interested in

finding out whether a new medication
Steps in Hypothesis Testing
will have any undesirable side effects.
By PennState, Eberly College of Science
https://online.stat.psu.edu/stat500/lesson/6a/6a. The researcher is particularly
2 concerned with the effect of the
medication on the patient’s pulse rate.
VIDEO WATCHING
The mean pulse rate is 82 beats per
Examples of Null and Alternative Hypotheses minute.
By Khan Academy
https://www.khanacademy.org/math/ap- 3. A teacher claims that there is
statistics/tests-significance-ap/idea-significance- significant relationship between
tests/v/examples-of-null-and-alternative-
hypotheses students’ academic achievement and
absenteeism.
4. The proportion of patients with heart
diseases is higher among smokers than
Practice Makes Perfect. Try to answer the
non-smokers.
activity.
B. Answer the following questions.
A. State the null and alternative hypothesis for
each situation and determine the type of test 1. State the steps of hypothesis in your
(one-tailed test or two-tailed test) required. own words.
1. An engineer hypothesizes that the mean
2. What does 𝛼 = 0.05 mean?
number of defects can be decreased in a
3. How can decision making in hypothesis
manufacturing process of USB drives by
testing be likened to a court trial as far
using robots instead of humans for
as defendant is concerned?
certain task. The mean number of
defective drives per 1,000 is 18. 4. When should a one-tailed test be used?
2. A psychologist feels that playing soft A two-tailed test?
music during a test will change the
5. Explain the difference between type I
results of the test.
error and type II error.

HOLY ANGEL UNIVERSITY I SCHOOL OF ARTS AND SCIENCES 11

 MATHEMATICS IN THE MODERN WORLD

2.2.2. STATISTICAL TESTS For example, IQ scores are said to be

normally distributed as illustrated in the figure
In this topic, commonly used statistical
below. The mean, median, and mode of IQ
tests will be explored. Before we begin, let us
scores is 100. This means that 50% of the
have a review of an important concept related
population have an IQ score below 100, while
to statistical test, that is, normal distribution.
the other 50% have an IQ score above 100.
The assumption on normality of a variable will
Moreover, 68% of the population have an IQ
also determine the test statistic to be employed.
score between 85 and 115.
How do we know if a variable is normally
distributed? Can you name some variables
which are said to be normally distributed?

Many continuous variables such as

height and IQ score are approximately normally
distributed. A normal distribution is a
probability distribution whose graph is
continuous, bell-shaped, and symmetric. Table
2.5 shows the properties of a normal
distribution. Source: https://www.iq-test-for-kids.com

Some Properties of a Normal Distribution Figure 2.6 IQ Score Normal Distribution Curve
▪ A normal distribution curve is bell-shaped.
▪ The mean, mode, and median are equal and located
at the center of distribution. Suppose the mean IQ sore of college
▪ A normal distribution has only one mode. freshmen of Holy Angel University is 110. Is the
▪ A normal distribution curve is symmetric about the
mean. mean IQ score of first year college students of
▪ A normal distribution curve is continuous and never Holy Angel University significantly higher than
touches the x-axis.
▪ The total area under a normal distribution is equal the population mean IQ score? To answer this
to 1 or 100%.
question, we need to perform statistical
▪ The area under the part of a normal distribution
curve that lies within 1 standard deviation of the hypothesis test using appropriate test statistic
mean is approximately 0.68, within 2 standard
deviations, about 0.95, and within 3 standard and applying the concept of normal
deviations, about 0.997. distribution.
Table 2.5 Properties of a Normal Distribution

HOLY ANGEL UNIVERSITY I SCHOOL OF ARTS AND SCIENCES 12

 MATHEMATICS IN THE MODERN WORLD

Which Statistical Test to Use? Level of Mutually Rank Equidistant Meaningful

Measurement Exclusive Order Values/
Zero Point
Categories Difference
The use of statistical test depends on Nominal Yes

many assumptions such as the normality of Ordinal Yes Yes

Interval Yes Yes Yes
distribution of data, scale of measurement of
Ratio Yes Yes Yes Yes
variables involved, and the hypotheses being
Table 2.6 Characteristics of Level of Measurement
tested. For example, a researcher wants to
determine whether there is a relationship Using the table, let us identify the level of
between the college students’ body mass index measurement of temperature. Temperature has
and mathematics achievement. If we assume two mutually exclusive categories: hot and cold.
that body mass index and mathematics Temperature may be ranked order as extremely
achievement are normally distributed and the hot, very hot, and hot. Temperature has
dare the in ratio scale, we can use Pearson equidistant values, for example, 38 oC is 2 oC higher
product moment correlation or Pearson’s r to than 36 oC. Temperature has no meaningful and
determine the strength of the relationship absolute zero. 0 oC does not mean no temperature
between body mass index and mathematics at all. Therefore, temperature is classified as
achievement among college students. interval-scale.
Knowing the scales of measurement of Practice Makes Perfect. What is the level of
variables is also important in conducting measurement of each variable?
statistical tests. Table 2.6 shows the
a. IQ scores
characteristics of the four levels or scales of
b. colors of athletic shirts
measurement (Bluman, 2016). A variable or
c. weight
data may have one or more level of
d. income
measurement. There are four characteristics to
e. religious affiliation
determine the highest scale of a variable. Does
f. letter grades (A+, A, A-, B+, …)
the variable have mutually exclusive
g. time of curing period
categories? Can the variable be ranked order?
h. military ranks
How much greater or less a data value is than
i. Likert scale for stress
another? Does the variable represent an
j. test score
absolute and meaningful zero?

HOLY ANGEL UNIVERSITY I SCHOOL OF ARTS AND SCIENCES 13

 MATHEMATICS IN THE MODERN WORLD

There is a wide range of statistical tests. Situation: A teacher wishes to see there is a difference in the mathematics
achievement of two groups of students who are exposed on different learning
As mentioned before, the decision of which to environments. One group of students is exposed to an online learning
environment, while the other group is exposed to a face-to-face classroom
environment. The former group has 40 students, while the latter group has 38
statistical test to use depends on the students. The teacher administers a 50-item posttest to measure students’
mathematics achievement to both groups. The mean test score of online
distribution of data, the type of variable, and learning environment groups is 38.5 and the standard deviation is 6.2. The mean
test score of face-to-face classroom environment group is 35.6 and the standard
the purpose of the hypothesis testing. Below is deviation is 4.5. Assume that the variable is normally distributed, can the teacher
conclude that there is difference in mathematics achievement of the two groups
list of just a few common statistical tests. An of students at level of significance of 5%?
illustrative example (Table 2.7) of the use of Step Procedure
1 Ho: There is no difference in the mathematics achievement of
statistical tests vis-à-vis hypothesis testing. the two groups of students. In symbol, ̅̅̅
𝑥1 = ̅̅̅.
𝑥2
Ha: There is difference in the mathematics achievement of the
two groups of students. In symbol, ̅̅̅ 𝑥2 (This is the claim
𝑥1 ≠ ̅̅̅.
Parametric Tests of the teacher.)
2 The level of significance is set at 5% or 0.05, in symbol, 𝛼 =
0.05.
▪ Z-test for one sample mean The hypothesis testing requires non-directional test or two-
tailed test because the alternative hypothesis makes use of
▪ T-test for one sample mean the symbol ≠.
3 𝑥1 = 38.5
̅̅̅ 𝑥2 = 35.6
̅̅̅
▪ Paired t-test 𝑠1 = 6.2 𝑠2= 4.5
𝑛1 = 40 𝑛2= 38
▪ Independent t-test Assume 𝜇1 − 𝜇2 = 0.
4 The hypothesis requires testing the difference between two
▪ Pearson product moment correlation means of independent samples. The appropriate statistic test
is t-test of two means for independent samples because the
coefficient or Pearson’s r sample data are independent of one another and the
population standard deviation is not known. As mentioned, the
▪ Analysis of variance (ANOVA) population is assumed to be normally distributed.

▪ Linear regression Formula of t statistic: (Variances are unequal.)

(𝑥 𝑥2 − (𝜇1 − 𝜇2)
̅̅̅1 − ̅̅̅)
Nonparametric Tests 𝑡=
(𝑠 )2 (𝑠 )2
√ 1 + 2
𝑛1 𝑛1
▪ Chi-square test (38.5 − 35.6) − (0 − 0)
𝑡=
2 2
▪ Spearman rank correlation √(6.2) + (4.5)
40 38
▪ Wilcoxon sign-rank test 𝑡 ≈ 2.373

▪ Sign test Degrees of freedom = smaller value of 𝑛1 − 1 or 𝑛2 − 1.

𝑛1 − 1 = 40 – 1 = 39
𝑛1 − 1 = 38 – 1 = 37
Note: Nonparametric tests are used when the Since 37 is smaller than 39, the degrees of freedom is 37.
Further Reading
data is not normally distributed. 5 The computed value (t) is 2.373.
The tabular value at 𝛼 = 0.05 and two-tailed test is 2.026
(based on Student t distribution table).
Types of Statistical Tests Since, the computed value is greater than the tabular value
By University of Minnesota (2.373 > 2.026), we accept the null hypothesis.
6 Therefore, there is not enough evidence to support the claim
https://cyfar.org/types-statistical-tests that there is difference in the mathematics achievement of two
groups of students.
Table 2.7 Example of Hypothesis Testing using T-Test

HOLY ANGEL UNIVERSITY I SCHOOL OF ARTS AND SCIENCES 14

 MATHEMATICS IN THE MODERN WORLD

Activity Supplemental Readings

Normal Distribution
Hypothesis testing is a formal https://www.dummies.com/education/math/statistics
procedure for investigating our ideas about /understanding-the-statistical-properties-of-the-
normal-distribution/
the world using statistics. Statistical tests are
Hypothesis Testing Examples
used in hypothesis testing. Choosing the right https://online.stat.psu.edu/statprogram/reviews/stati
statistical tests will make our hypothesis stical-concepts/hypothesis-testing/examples

testing valid. In this activity, you must know Videos

the uses of various statistical tests for
How to Perform Pearson Correlation Test in Excel
comparison, relationship, and regression and https://www.youtube.com/watch?v=Ev86DMtLXOk

provide illustrative examples for each Chi Squared Test

statistical test. Complete the table below by https://www.youtube.com/watch?v=qYOMO83Z1WU&
t=2s
identifying the conditions required and uses of
each statistical test and give a situation where Choosing which Statistical Test to Use – Statistics Help
https://www.youtube.com/watch?v=rulIUAN0U3w
it can be applied.

Statistical Conditions References

Examples
Tests and Uses
z-test Bluman, A. (2016). Elementary statistics: A step by
step approach, A brief Version (7th ed.).
Paired t-test McGraw-Hill Education.
Gonzales, J. (2015). Essential Statistics. MaxCor
Independent t-
Publishing.
tests
Chapter 8: Introduction to Hypothesis Testing.
ANOVA https://www.sagepub.com/sites/default/files
/upm-binaries/40007_Chapter8.pdf
Pearson r
Rosenthal, G., & Rosenthal, J. (2012). Statistics and
Spearman rho Data Presentation for Social Work (Par II
Inferential Statistics and Data Presentation).
Chi-square http://web.a.ebscohost.com/ehost/ebookvie
wer/ebook/ZTAwMHh3d19fNDMyMjc1X19B
Tg2?sid=679681db-61d8-4f59-8127-
f3ec8727e9c5@sdc-v-
sessmgr01&vid=5&format=EB&rid=1

HOLY ANGEL UNIVERSITY I SCHOOL OF ARTS AND SCIENCES 15