STUDY GUIDE

STATISTICS & PROBABILITY | UNIT 20

Chi-Square Tests

Table of Contents

Introduction .................................................................................................................................... 3
Test Your Prerequisite Skills ........................................................................................................ 4
Objectives ...................................................................................................................................... 4
Lesson 1: Introduction to Chi-Square Tests
- Warm Up! ........................................................................................................................... 5
- Learn about It! ................................................................................................................... 6
- Let’s Practice! ..................................................................................................................... 7
- Check Your Understanding! ............................................................................................ 10
Lesson 2: Chi-Square Distribution
- Warm Up! ......................................................................................................................... 12
- Learn about It! ................................................................................................................. 12
- Let’s Practice! ................................................................................................................... 14
- Check Your Understanding! ............................................................................................ 17
Lesson 3: Goodness of Fit Test
- Warm Up! ......................................................................................................................... 19
- Learn about It! ................................................................................................................. 19
- Let’s Practice! ................................................................................................................... 20
- Check Your Understanding! ............................................................................................ 25
Lesson 4: Chi-Square Test for Independence
- Warm Up! ......................................................................................................................... 27
- Learn about It! ................................................................................................................. 27
- Let’s Practice! ................................................................................................................... 29
1

Copyright © 2018 Quipper Limited

 STUDY GUIDE

- Check Your Understanding! ............................................................................................ 35

Challenge Yourself! ..................................................................................................................... 36
Performance Task ....................................................................................................................... 37
Wrap-up ....................................................................................................................................... 39
Key to Let’s Practice! .................................................................................................................... 39
References ................................................................................................................................... 41

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Click Home icon to go back to

Table of Contents
GRADE 11/12 | STATISTICS & PROBABILITY

UNIT 20

Chi-Square Tests
In daily life, relationships between our status and our decisions are
often observed. Our perception towards a particular topic or event
may have been influenced by groups in which we belong. For
example, as we grow older, we tend to see things in a different
perspective than a younger individual does. These relationships
between our choices and our status are often subtle.

Differences on choices can be established as existent or

nonexistent by using chi-square test. In particular, chi-
square test could help us determine what possible
factors affect someone’s preference or predisposition.
For instance, determining whether gender is related to
anxiety or overspending gives us an idea on which
group of people should be addressed or prioritized by
intervention programs.

In this unit, we will explore how chi-square test can be applied to different examples such
as the ones stated above and what other kinds of problem it can solve.

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Test Your Prerequisite Skills

 Determining whether variables are qualitative or quantitative

 Calculating the probability of a given event

Before you get started, answer the following items on a separate sheet of paper. This will
help you assess your prior knowledge and practice some skills that you will need in
studying the lessons in this unit. Show your complete solution.

1. Determine whether each of the following variables is qualitative or quantitative.

a. age d. gender
b. civil status e. educational attainment
c. weight

2. Determine the probability of each of the following events.

a. Getting an even number when a die is rolled
b. Picking a red marble from a bowl with 10 green marbles, 5 red marbles, and 10
blue marbles
c. Getting a heart card from a deck of cards
d. Getting a face card from a deck of cards
e. Getting a number less than 3 when a die is rolled

Objectives

At the end of this unit, you should be able to

 compute the chi-square statistic;
 illustrate the chi-square distribution; and
 solve problems involving chi-square tests.

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Lesson 1: Introduction to Chi-Square Tests

Warm Up!

What’s Your Preference?

Materials Needed: pen and paper

Instructions:
1. This activity should be done by the whole class.

2. The teacher asks the following questions aloud:

a. Who among you likes the color pink?
b. Who among you prefers to work abroad?
c. Who among you always sleeps late?
d. Who among you is addicted to online games?

3. For each question, the teacher will note the number of responses using the
following table:

Male Female
Question a
Question b
Question c
Question d

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Learn about It!

In Warm Up!, you may have noticed that some questions got more responses from girls
and some got more responses from boys. In that sense, we could say that gender may
probably have a relationship to color preference, for example. This relationship can be
tested using chi-square test.

Definition 1.1: A chi-square test is any hypothesis test which assumes

that the sampling distribution of the test statistic follows
the chi-square distribution. It is mostly used to test
relationships between qualitative variables.

There are mainly two types of chi-square test that will be discussed later in this unit: chi-
square goodness-of-fit test and chi-square test for independence.

Determining whether there exists a relationship between age bracket and political
candidate preference is an example of a study where chi-square test can be used.
Inquiries such as “Does gender affect food preference?” are also another example where
chi-square test can be applied. If gender, in the latter example, affects food preference,
we say that the two categorical variables—gender and food preference—are dependent.
Otherwise, we conclude that the variables involved are independent, which means that
there exists no relationship between the variables.

The chi-square statistic is given by

where and are the observed and expected frequencies of category , respectively.

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Let’s Practice!

Example 1: Suppose the expected frequencies of two given categories are 200 and 250.
During the study, the researcher observed 195 and 255 as the frequencies of
the two categories, respectively. What is the value of ?

Solution: Use the formula to solve for .

Try It Yourself!

Suppose the expected frequencies of two given categories are 120 and 80. During
the study, the researcher observed 105 and 95 as the frequencies of the two
categories, respectively. What is the value of ?

Example 2: The following table shows the expected frequencies and actual observed
frequencies of three categories in a study. Calculate .

Expected Frequency Observed Frequency

Category 1 90 85
Category 2 50 55
Category 3 10 10

Solution: Use the formula to solve for .

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Try It Yourself!

The following table shows the expected frequencies and actual observed
frequencies of three categories in a study. Calculate .

Expected Frequency Observed Frequency

Category 1 50 55
Category 2 40 25
Category 3 10 20

Example 3: The following table shows the expected frequencies and actual observed
frequencies of four categories in a study. Calculate .

Expected Frequency Observed Frequency

Category 1 100 75
Category 2 100 120
Category 3 100 110
Category 4 100 95

Solution: Use the formula to solve for .

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Try It Yourself!

Suppose the expected frequencies of four different categories in a given study are
equal. If the experiment is conducted 400 times and the actual observed
frequencies in the study are indicated below, calculate the critical value of .

Observed Frequency
Category 1 110
Category 2 70
Category 3 90
Category 4 130

More Real-World Problems

Example 4: A publishing company claims that 30% of

the books they publish are textbooks, 20%
are reference books, and 50% are
inspirational books. A random sample of
300 books consists of 85 textbooks, 62
reference books, and 153 inspirational books. What is the value of for this
problem?

Solution: Organize the given data using a table. Note that the expected frequency of a
category is equal to the percentage of the category multiplied by the sample
size.
Expected Frequency Observed Frequency
Textbooks
Reference Books
Inspirational Books
9

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Use the formula to compute for .

Try It Yourself!

A machine is supposed to deal cards randomly. You

counted 100 cards and observed the following
frequencies:

Suit Frequency
Spade 26
Heart 22
Diamond 24
Club 28

Calculate .

Check Your Understanding!

1. Determine whether each of the following statements is true or false.

a. The value of the chi-square statistic cannot be negative.
b. The sum of the observed frequencies in a chi-square test is equal to the sample
size.
c. The observed frequencies in a chi-square test need not be equal for all
categories.
d. The expected frequencies in a chi-square test are equal for all categories.
10

Copyright © 2018 Quipper Limited

 STUDY GUIDE

2. Solve for the chi-square statistic given the following data.

a.
Expected Frequency Observed Frequency
Category 1 50 44
Category 2 50 56

b.
Expected Frequency Observed Frequency
Category 1 75 68
Category 2 100 102
Category 3 25 30

c.
Expected Frequency Observed Frequency
Category 1 200 150
Category 2 200 250
Category 3 200 212
Category 4 200 188

11

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Lesson 2: Chi-Square Distribution

Warm Up!

Conduct a Poll

Materials Needed: pen and paper

Instructions:
1. This activity should be done in groups of ten.
2. In each group, members respond to the following questions by saying Yes or No:
 Do you play online games?
 Can you play the guitar?
 Can you sing?
 Do you study prefer studying late at night?
 Do you attend any fitness activity?
3. One group member takes note of the number of Yes and No responses to each
question and whether each response is from a male or a female. Then he/she
tabulates the frequencies. Compare results with other groups.

Learn about It!

In Warm Up!, you may have observed that although questions for all groups are the same,
the proportion of boys or girls who responded yes vary. In one group, boys could be the
majority who responded Yes to a question but in another group, the majority of the
responses for the same question may have come from the girls. Using chi-square test, the
independence of gender and any other qualitative variable can be determined. In this
lesson, we will discuss the chi-square distribution.

12

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Similar to the normal distribution and binomial distribution, the chi-square distribution (or
distribution) is a continuous probability distribution. In the chi-square distribution, the
minimum value of the random variable is 0 and there is no maximum value.

A degree of freedom is associated with each distribution. The following table shows the
critical values of the distribution with 1 to 15 degrees of freedom given the right tail
area.

Right-Tail Area
df
0.995 0.975 0.9 0.5 0.1 0.05 0.025 0.01 0.005
1 0.000 0.001 0.016 0.455 2.706 3.841 5.024 6.635 7.879
2 0.010 0.051 0.211 1.386 4.605 5.991 7.378 9.210 10.597
3 0.072 0.216 0.584 2.366 6.251 7.815 9.348 11.345 12.838
4 0.207 0.484 1.064 3.357 7.779 9.488 11.143 13.277 14.860
5 0.412 0.831 1.610 4.351 9.236 11.070 12.833 15.086 16.750
6 0.676 1.237 2.204 5.348 10.645 12.592 14.449 16.812 18.548
7 0.989 1.690 2.833 6.346 12.017 14.067 16.013 18.475 20.278
8 1.344 2.180 3.490 7.344 13.362 15.507 17.535 20.090 21.955
9 1.735 2.700 4.168 8.343 14.684 16.919 19.023 21.666 23.589
10 2.156 3.247 4.865 9.342 15.987 18.307 20.483 23.209 25.188
11 2.603 3.816 5.578 10.341 17.275 19.675 21.920 24.725 26.757
12 3.074 4.404 6.304 11.340 18.549 21.026 23.337 26.217 28.300
13 3.565 5.009 7.042 12.340 19.812 22.362 24.736 27.688 29.819
14 4.075 5.629 7.790 13.339 21.064 23.685 26.119 29.141 31.319
15 4.601 6.262 8.547 14.339 22.307 24.996 27.488 30.578 32.801

To use the chi-square table to determine a critical value, the degrees of freedom and
significance level are needed. Moreover, to estimate a -value, and -statistic are
needed.

In summary, the chi-square distribution has the following general characteristics:

 The chi-square distribution is a continuous distribution with the random variable

ranging from 0 to infinity in the positive direction. This means that we can never
assume a negative value of .
13

Copyright © 2018 Quipper Limited

 STUDY GUIDE

This is evident in the given formula of the chi-square statistic wherein the
differences between the observed frequencies and the expected frequencies are
squared. Hence, is can only be positive or zero.

 The shape of the chi-square distribution depends on the degrees of freedom .

When is small, the shape of the curve tends to be skewed to the right. But as
becomes larger, the shape becomes more bell-shaped and can then be
approximated using the normal distribution.
Probability

 The mean of a chi-square distribution is equal to its degrees of freedom while the
variance is twice its degrees of freedom.

Let’s Practice!

Example 1: What is the critical value in a chi-square test if its distribution has 11
degrees of freedom and 0.1 significance level?

Solution: Using the chi-square table with and , we have as critical

value.
14

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Try It Yourself!

What is the critical value in a chi-square test if its distribution has 5 degrees of
freedom and 0.025 significance level?

Example 2: What is the -value of if its distribution has 2 degrees of

freedom?

Solution: Using the -table with and as critical value, we have

.

Try It Yourself!

What is the -value of if its distribution has 7 degrees of freedom?

Example 3: The chi-square statistic of a study on categorical data is 12.2 with 5

degrees of freedom. What is the -value of ?

Solution: Since with , using the -table, we have

but . Thus, the -value of is between 0.025 and 0.05.

Try It Yourself!

The chi-square statistic of a study on categorical data is with 10 degrees of

freedom. What is the -value of ?

15

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Real-World Problems

Example 4: A machine is supposed to deal cards randomly. In an

experiment, you were dealt cards with the
following suits:

Suit Frequency
Club 524
Heart 520
Diamond 498
Spade 458

Calculate the chi-square statistic of the given data and its -value. Note that
the degrees of freedom of the given data is equal to , where is the
number of categories.

Solution: Since , . Moreover, since the machine is expected

to deal cards randomly, the probability of obtaining each type of card must
be 0.25. Hence, the expected frequency of each suit is

Use the formula to solve for .

Since with , using the -table, we have but

. Thus, the -value of is between and 0.5.

16

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Try It Yourself!

A machine is supposed to deal cards randomly. In an

experiment, you were dealt 150 cards with the following
suits:
Suit Frequency
Spade 30
Heart 22
Diamond 55
Club 43

Calculate the chi-square statistic of the given data and its -value.

Check Your Understanding!

1. State whether each of the following statements is true or false.

a. In a chi-square test, as the -statistic increases, its -value also increases.
b. The sum of the expected frequencies in a chi-square test is equal to the sample
size.
c. As the degree of freedom of a chi-square distribution increases, its graph
approaches the normal distribution.
d. The variance of a chi-square distribution is equal to its degrees of freedom.

2. Determine the -value given the following statistics and degrees of freedom.
a. ,
b. ,
c. ,
d. ,
e. ,

17

Copyright © 2018 Quipper Limited

 STUDY GUIDE

3. Determine the critical value in a chi-square test given the following significance
levels and degrees of freedom.
a.
b.
c.
d.
e.

4. Calculate the statistic given the following data.

Observed Frequency Expected Frequency

Category 1 212 200
Category 2 213 200
Category 3 198 200

18

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Lesson 3: Goodness of Fit Test

Warm Up!

Toss It!

Materials Needed: pen, paper, coin

Instructions:
1. Divide the class into four groups.
2. For each group, toss the coin 50 times and record the outcomes.
3. Each group should briefly present their results.

Learn about It!

In Warm Up!, each group may have recorded different sets of outcomes. If you compare
your outcomes with the theoretical probability of obtaining a head or a tail, that is, ,
your observed outcomes may not match this theoretical probability. It does not mean,
however, that the theoretical and experimental probabilities are significantly different. A
goodness-of-fit test can help us determine whether the outcomes that we observe match
the outcomes that we expect.

The chi-square test is can be employed to determine whether the observed frequencies
“fit” the expected frequencies, making a claim consistent, or to determine relationships
between categorical variables.

19

Copyright © 2018 Quipper Limited

 STUDY GUIDE

In a chi-square goodness-of-fit test, the null hypothesis states that each observed
frequency is equal to its expected frequency while the alternative hypothesis states that at
least one of the observed frequencies is different from its corresponding expected
frequency.

Note that the chi-square goodness-of-fit test is appropriate when the sampling method is
simple random sampling, the variables are categorical, and the expected frequency in
each category is at least 5.

The following are the steps in performing a chi-square goodness-of-fit test:

1. State the hypotheses.

2. Specify the level of significance.
3. Use the -table to determine the critical value and rejection region. Use ,
where is the number of categories/levels.
4. Solve for the statistic.
5. Determine whether the -statistic falls in the critical region. If it does, reject the
null hypothesis. If it does not, do not reject the null hypothesis.
6. Interpret the result based on the stated hypotheses.

Note that in most applications of the chi-square goodness-of-fit test, it is always assumed
that we are doing a one-tailed test.

Let’s Practice!

Example 1: Suppose a researcher conducted an experiment to determine if rats have an

exit preference by sending rats to a ramp that will lead to three doors of
different colors. State the hypotheses for this experiment.

20

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Solution: If rats have no preference, we expect that the three doors would be chosen
an equal number of times. Hence, the null hypothesis should state that the
probability of each door being chosen is .

The alternative hypothesis should state that rats have a door preference,
that is, the probability of at least one out of the three doors being chosen is
different from the others.

: At least one is not equal to .

Try It Yourself!

A man was observed buying breakfast in a food stall with four different choices. If
is the probability that the man chooses stall , what are the appropriate hypotheses
if a study would want to know if the man has a food preference in the stall?

Example 2: Suppose the experiment in Example 1 was conducted to 90 rats, whose

results are shown on the table below.

Door 1 Door 2 Door 3

Observed Frequency 20 39 31

Determine the degrees of freedom, expected frequencies, and rejection

region in a chi-square goodness-of-fit test using the given data. Use .

Solution: Given that , we have . Since the expected

probability of each door is , the expected frequencies are

21

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Using the chi-square table with and , we have a critical value

of . Thus, the rejection region contains all values of such that
.

Try It Yourself!

Suppose an experiment about color preference was conducted to 100 respondents,

whose results are shown on the table below.

Color 1 Color 2 Color 3 Color 4

Observed Frequency 20 30 27 23

Determine the degrees of freedom, expected frequencies, and rejection region in a

chi-square goodness-of-fit test using the given data. Use .

Example 3: Suppose the experiment in Example 1 was conducted to 90 rats, whose

results are shown on the table below.

Door 1 Door 2 Door 3

Observed Frequency 20 39 31

Is there enough evidence to conclude that the rats have a door preference?

Solution: Use the formula to solve for .

22

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Since , we reject the null hypothesis. Thus, at least one is

not equal to . Hence, there is enough evidence to conclude that the rats
have a door preference.

Try It Yourself!

A company claims that they produce equal numbers of the four different variations
of their product. If a random sample of 200 of their products was taken and the
following frequencies were recorded, test the claim of the company using a 0.01
level of significance.

Variation Observed Frequency

Type 1 60
Type 2 55
Type 3 45
Type 4 40

More Real-World Problems

Example 4: A sports company prints basketball cards. It claims

that 40% of the cards are all-stars, 10% are rookies,
and 50% are veterans. A random sample of 100 cards
consists of 35 all-stars, 11 rookies, and 54 veterans.
Use a level of significance of 0.05 to determine
whether the claim of the company is true.

Solution:

Step 1: State the hypotheses.

23

Copyright © 2018 Quipper Limited

 STUDY GUIDE

: The percentage of all-star, rookie, and veteran cards are 40%, 10%, and
50%, respectively.

: At least one of the percentages in is false.

Step 2: Specify the level of significance.

It is given that .

Step 3: Use the -table to determine the critical value and rejection region.

Since , . Using the -table with and

, we have 5.991 as critical value. Hence, the rejection region contains
all values of such that

Step 4: Solve for the statistic.

Use the percentage of each category to determine their expected

frequencies. Then use a table to organize the expected and observed
frequencies.

Expected Frequency Observed Frequency

All-Star
Rookie
Veteran

Use the formula to solve for .

24

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Step 5: Determine whether the -statistic falls in the critical region.

Since , we fail to reject the null hypothesis.

Step 6: Interpret the result based on the stated hypotheses.

There is not enough evidence to say that the company’s claim on the
percentages of basketball card designs is not true.

Try It Yourself!

A publishing company claims that 30% of the books

they publish are textbooks, 20% are reference
books, and 50% are inspirational books. A random
sample of 300 books consists of 85 textbooks, 62
reference books, and 153 inspirational books. Use a
0.05 level of significance to determine whether the
claim of the company is true.

Check Your Understanding!

1. Determine whether each of the following statements is true or false.
a. The goodness-of-fit test uses the chi-square distribution.
b. If the -value of the chi-square critical value is less than the level of significance,
we do not reject the null hypothesis.
c. The null hypothesis of a goodness-of-fit test states that at least one of the
expected frequencies does not match the observed frequency.
d. The in a goodness-of-fit test is equal to the number of categories.

25

Copyright © 2018 Quipper Limited

 STUDY GUIDE

2. A company’s food product has three different flavors. The company claims that
they distribute equal numbers of each flavor to their partner stores. In a sample of
900 products from a partner store, the following frequencies are observed:

Flavor Observed Frequency

Flavor 1 295
Flavor 2 320
Flavor 3 285

Is there enough evidence to say that the partner store does not receive all flavors
equally?

3. It is reported in an old survey that the proportions of blood types A, B, AB, and O in
the country are 0.40, 0.10, 0.20, and 0.30, respectively. To determine whether the
present population still fits this reported percentage, 200 samples are selected.
Given the following observed frequencies, test if the reported percentages are still
true.

Blood Type Observed Frequency

A 75
B 22
AB 45
O 58

26

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Lesson 4: Chi-Square Test for Independence

Warm Up!

Book Preference

Materials Needed: pen and paper

Instructions:
1. This activity should be done as a class.
2. Each student should choose one book preference among manga, romantic
novel, and science fiction. Then tabulate the results using the following table:

Manga Romantic Novel Science Fiction

Male
Female

3. What can you infer from the table?

Learn about It!

In Warm Up!, you may have observed that there seems to be a relationship between
gender and book preference. However, we cannot statistically conclude by merely looking
at the table. There might be an observable difference between the responses of male and
female but we have to make sure that the difference is significant. The chi-square test for
independence allows us to arrive at such conclusion.

27

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Using the chi-square test for independence, we can conclude whether two categorical
variables are dependent or independent of each other. That is, we can answer the
question: “Does one variable affect the other?”

The chi-square test for independence can be used when there exist two categorical
variables from a single population. It is used to determine if there is a significant
association between these categorical variables.

In a chi-square test for independence, the null hypothesis states that the two variables
under study are independent. On the other hand, the alternative hypothesis states that
the two variables are dependent.

The chi-square test for independence can be used when the sampling method is simple
random sampling, the variables are categorical, and the expected frequency of each entry
in the table is at least 5. Also, the degrees of freedom is given by

where is the number of levels for the first variable and is the number of levels for the
second variable.

The expected frequencies are computed separately for each category of the first variable
at each category of the second variable. The expected frequency at category of the
first variable and category of the second variable is defined by

where is the total number of sample observations in category of the first variable, is
the total number of sample observations in category of the second variable, and is the
sample size.

28

Copyright © 2018 Quipper Limited

 STUDY GUIDE

The -statistic is defined by

where and are the observed and expected frequencies, respectively, at category
of the first variable and category of the second variable.

Similar to the chi-square goodness-of-fit test, the following are the steps in performing a
chi-square test for independence:

1. State the hypotheses.

2. Specify the level of significance.
3. Use the -table to determine the critical value and rejection region.
4. Solve for the statistic.
5. Determine whether the -statistic falls in the critical region. If it does, reject the
null hypothesis. If it does not, do not reject the null hypothesis.
6. Interpret the result based on the stated hypotheses.

Let’s Practice!

Example 1: Suppose a study about the relationship between gender and type of pet
owned is to be conducted. What are the appropriate hypotheses for this
study?

Solution: Since the two variable under study are categorical, we can use a chi-square
test for independence. Thus, we have the following hypotheses:

: Gender and type of pet owned are independent.

: Gender and type of pet owned are dependent.
29

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Try It Yourself!

Suppose a study about the relationship between citizenship and food preference is
to be conducted. What are the appropriate hypotheses for this study?

Example 2: A survey was conducted to determine the effectiveness of a two-shot flu

vaccine. The following table shows observed frequencies:

No Vaccine One Shot Two Shots Total

Flu 24 9 13 46
No Flu 289 100 565 954
Total 313 109 578 1000

If a chi-square test for independence is to be performed, determine the

degrees of freedom and the expected frequency of each category.

Solution: Since there are categories (flu, no flu) for the first variable and
categories (no vaccine, one shot, two shots) for the second variable,
.

Use the formula to solve for the expected frequencies.

30

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Try It Yourself!

A survey was conducted to determine whether performance in mathematics is

dependent on gender. The following table shows the observed frequencies:

Low Average High

Total
Performance Performance Performance
Male 160 280 90 530
Female 118 270 82 470
Total 278 550 172 1000

If a chi-square test for independence is to be performed, determine the degrees of

freedom and the expected frequency of each category.

Example 3: In the given problem in Example 2, what is the value of the -statistic and
what is the appropriate conclusion? Use .

Solution: We have the following hypotheses:

: Presence of flu and number of flu shots are independent.

: Presence of flu and number of flu shots are dependent.

Using the -table with and , we have as critical value.

Thus, the rejection region contains all values of such that .

Use the formula to solve for .

31

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Since , we reject the null hypothesis. Thus, presence of flu

and number of flu shots are dependent.

Try It Yourself!

A survey was conducted to determine whether performance in mathematics is

dependent on gender. The following table shows the observed frequencies:

Low Average High

Total
Performance Performance Performance
Male 160 280 90 530
Female 118 270 82 470
Total 278 550 172 1000

Using 0.05 level of significance, is there enough evidence to say that performance in
mathematics is dependent on gender?

32

Copyright © 2018 Quipper Limited

 STUDY GUIDE

More Real-World Problems

Example 4: A poll surveyed a simple random sample of 100 people.

Respondents were classified by their gender and color
preference.

Green Yellow Purple Total

Male 23 11 7 41
Female 19 17 23 59
Total 42 28 30 100

Perform a chi-square test for independence to determine whether color

preference is dependent on gender. Use .

Solution:

Step 1: State the hypotheses.

: Gender and color preference are independent.

: Gender and color preference are dependent.

Step 2: Specify the level of significance.

It is given that .

Step 3: Use the chi-square table to determine the critical value and rejection region.

Since there are genders and color preference choices,

. Using the -table with and
, we have 5.991 as critical value. Thus, the rejection region contains
all values of such that .
33

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Step 4: Solve for the -statistic.

Use the formula to solve for the expected frequencies.

Use the formula to solve for .

Step 5: Determine whether the -statistic falls in the critical region.

Since , we reject the null hypothesis.

Step 6: Interpret the result based on the stated hypotheses.

There is enough evidence to say that color preference is dependent on

gender.

34

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Try It Yourself!

Determine whether there is a relationship between gender and sports preference

using the following data. Use 0.025 level of significance.

Basketball Badminton Volleyball Total

Male 121 45 36 202
Female 52 94 102 248
Total 173 139 138 450

Check Your Understanding!

1. Determine whether each of the following statements is true or false.

a. Chi-square test for independence can be applied to quantitative variables.
b. Chi-square test for independence can be used to determine potential
relationships between categorical variables.
c. The degrees of freedom in a chi-square test for independence is the product of
the number of categories of the two variables.
d. The null hypothesis of a chi-square test for independence states that there
exists a relationship between two categorical variables.
e. The alternative hypothesis of a chi-square test for independence states that the
two categorical variables are dependent.

35

Copyright © 2018 Quipper Limited

 STUDY GUIDE

2. Perform a chi-square test for independence given the following observed

frequencies. Use .
a.
Fiction Book Non-Fiction Book
Male 120 80
Female 90 130
b.
Amount Spent on Luxury
Less than Half More than Half
None
of Allowance of Allowance
Full-Time
210 25 3
Student
Working
180 20 1
Student

Challenge Yourself!

1. Why is the chi-square critical value always nonnegative?

2. Why is the calculation of the degrees of freedom in a chi-square goodness-of-fit test
different from that in a chi-square test for independence?
3. What would be the result of a chi-square goodness-of-fit test if the observed and
expected frequencies are equal in all categories? What is its implication?
4. What would be the result of a chi-square test for independence if the observed and
expected frequencies are equal in all categories? What is its implication?

36

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Performance Task

An educational institution plans to conduct a seminar about robotics. However, because

of limited resources, they can only accommodate a limited number of participants. The
institution wants to prioritize students who are interested in pursuing a science or
mathematics field in the future.

As one of the members of the organizing committee, you are tasked to survey 100
students about their preferred field in college (e.g. business, science, math, arts). In
general, you need to determine whether the field of interest of the students is dependent
on gender. This is to help the organizing committee in deciding which group to orient first.
You must organize your data and present it to the committee.

Performance Task Rubric

Below Needs Successful Exemplary

Criteria Expectation Improvement Performance Performance
(0–49%) (50–74%) (75–99%) (100%)
All
There is a There are only
computations
significant a few errors in All
are correct
Accuracy of number of the computations
and with
Computation errors in the computation are correct,
complete
and Analysis computations but there is no and the data
solution. The
of Data that led to clear basis in are analyzed
data are
wrong analysis the analysis of properly.
analyzed with
of data. data.
clear basis.

37

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Data are
Data are Data are
organized
organized organized
properly but
Data are not properly but properly. All
Organization some
organized some information
of Data information
properly. necessary needed in the
needed in the
parts are analysis are
analysis are
missing. present.
present.
The
There is no The presentation
observable presentation shows an in-
The
comprehension shows some depth
presentation
of the problem observable understanding
shows an
and the understanding of all the
Mathematical observable
method being of the required
Justifications complete
presented. The required mathematical
and understanding
solution covers mathematical components
Presentation of the
none of the knowledge. of the task. All
required
mathematical Some of the solutions and
mathematical
components solutions used procedures
components.
required to are incorrectly are presented
solve the task. executed. clearly and
concisely.

38

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Wrap-up

Key Terms/Formulas

Key Term Description

A chi-square test used to determine whether the
Goodness-of-Fit test observed frequencies are consistent with the
expected frequencies.

A chi-square test used to determine if whether two

Test for Independence
categorical variables are independent.

Key to Let’s Practice!

Lesson 1
1. 4.688
2. 16.125
3. 20
4. 0.8

39

Copyright © 2018 Quipper Limited

 STUDY GUIDE

Lesson 2
1. 12.833
2. 0.5
3.
4. 16.88;

Lesson 3
1. :
: At least one is not equal to .
2. ; ; The rejection region contains all values of such
that .
3. Since , do not reject the null hypothesis. Thus, there is not enough
evidence to refute the claim of the company.
4. Since , do not reject the null hypothesis. Thus, there is not
enough evidence to refute the claim of the publishing company about the
percentages of their books. Also, its p-value is between 0.5 and 0.9. Hence, at 0.05
level of significance, we can conclude that the company’s claim is consistent with
the sample data.

Lesson 4
1. : Citizenship and food preference are independent.
: Citizenship and food preference are dependent.
2. ; , , , , , and
.147.34, 91.16130.66258.580.84
3. Since , we fail to reject the null hypothesis. Thus, there is NOT
enough evidence to say that performance in mathematics is dependent on gender.
4. Since , we reject the null hypothesis. Thus, there is sufficient
evidence to say that gender and sports preference are dependent.

40

Copyright © 2018 Quipper Limited

 STUDY GUIDE

References

Math Is Fun. “Chi-Square Test”. Accessed October 26, 2018.

https://www.mathsisfun.com/data/chi-square-test.html

Mendenhall III, WIliam, Robert J. Beaver, and Barabara M. Beaver. Introduction to

Probability and Statistics. United States of America: Brooks/Cole Cengage Learning,
2013.

Statistics How To. “Chi-Square Test”. Accessed October 26, 2018.

https://www.statisticshowto.datasciencecentral.com/probability-and-statistics/chi-
square/

StatTrek. “Goodness-of-Fit test”. Accessed October 26, 2018. https://stattrek.com/chi-

square-test/goodness-of-fit.aspx

41

Copyright © 2018 Quipper Limited