by:

LESSER A. LOC, MAEd

Project FREE - Paglaum
 Supposing a soap
But one of the consumers
manufacturer is confident
claim that it DOES NOT.
that their product can kill
99.9% of skin germs

Our soap is
tested to kill No! You are
99.9% of skin lying.
germs
But the question is:

WHO IS TELLING THE TRUTH?

Which statement is true and
which is FALSE?
Which statement should be ACCEPTED and
which should be REJECTED?
 The statements made by both the
MANUFACTURER and the CONSUMER is
referred to as the:

HYPOTHESIS
To find out which hypothesis is true, we
will use perform the process known as:
HYPOTHESIS TESTING
A hypothesis test is a process that uses sample statistics to
test a claim about the value of a population parameter.
 HYPOTHESIS
A statistical hypothesis is an inference about a population
parameter. This inference may or may not be true. The
only sure way of finding the truth or falsity of a hypothesis
is by examining the entire population. Because this is not
always feasible, a sample is instead examined for the
purpose of drawing conclusions.

Hypothesis in statistics can be classified into two:

Null hypothesis (Ho) – a statement about a population based on past

observation or experimentation or simply a prevailing belief about the
population in consideration. This also refers to things are happening as
expected. Currently accepted value or parameter.

Alternative (Ha) – A statement that challenges the null hypothesis which

usually contains the claim to be tested or investigated.
 How to write the
Null Hypothesis
The null hypothesis (Ho) must contain the
condition of equality and can be written in
the following ways:
𝐻𝑜: 𝜇 = 𝑘 𝐻𝑜: 𝜇 ≤ 𝑘 𝐻𝑜: 𝜇 ≥ 𝑘

Word clues: equal to, at least (greater than

or equal to), at most (less than or equal to)

Note: k could be any value

 How to write the
Alternative Hypothesis
The alternative hypothesis (Ha) is a statement that
challenges the null hypothesis. Hence, the null
hypothesis must contain the condition of
inequality. For the mean, the alternative hypothesis
can be stated in only one of these three possible
forms:

𝐻𝑎: 𝜇 ≠ 𝑘 𝐻𝑎: 𝜇 > 𝑘 𝐻𝑎: 𝜇 < 𝑘

Word clues: different, greater than, less than

Note: k could be any value

 EXAMPLE 1
It is believed that a candy machine makes
chocolate bars that are on average 5g. A worker
claims that the machine after maintenance no
longer makes 5g bars. Write Ho and Ha.
In words,
Remember!
Ho: C𝑎𝑛𝑑𝑦 𝑚𝑎𝑐ℎ𝑖𝑛𝑒𝑠 𝑚𝑎𝑘𝑒 𝑐ℎ𝑜𝑐𝑜𝑙𝑎𝑡𝑒 𝑏𝑎𝑟𝑠
𝑡ℎ𝑎𝑡 𝑎𝑟𝑒 𝑜𝑛 𝑎𝑣𝑒𝑟𝑎𝑔𝑒 5𝑔. The first statement shows
EQUALITY. This must be
Ha: C𝑎𝑛𝑑𝑦 𝑚𝑎𝑐ℎ𝑖𝑛𝑒𝑠 𝑑𝑜 𝑛𝑜𝑡 𝑚𝑎𝑘𝑒 5𝑔
chocolate bars. the NULL HYP.
In symbols,
The second statement
Ho: 𝜇 = 5𝑔 shows INEQUALITY. The
clue is the word “NO
Ha: 𝜇 ≠ 5𝑔 LONGER”
 EXAMPLE 2
Doctors believed that teenager sleeps on
average AT LEAST 8 hours per day. A
researcher believes that teens on average
sleep shorter. Write Ho and Ha.
In words, The word AT LEAST (greater
𝑇𝑒𝑒𝑛𝑎𝑔𝑒𝑟 𝑠𝑙𝑒𝑒𝑝𝑠 𝑜𝑛 𝑎𝑣𝑒𝑟𝑎𝑔𝑒 𝑎𝑡 𝑙𝑒𝑎𝑠𝑡 than or equal to) in the first
Ho: 8 ℎ𝑜𝑢𝑟𝑠 𝑝𝑒𝑟 𝑑𝑎𝑦. statement signifies a
condition of EQUALITY, so
Ha: 𝑇𝑒𝑒𝑛𝑎𝑔𝑒𝑟 𝑠𝑙𝑒𝑒𝑝𝑠 𝑠ℎ𝑜𝑟𝑡𝑒𝑟 𝑡ℎ𝑎𝑛
8 ℎ𝑜𝑢𝑟𝑠 𝑝𝑒𝑟 𝑑𝑎𝑦. the first statement must the
NULL HYPOTHESIS.
In symbols,
The second statement shows
Ho: 𝜇 ≥ 8 ℎ𝑜𝑢𝑟𝑠 INEQUALITY using the word

Ha: 𝜇 < 8 ℎ𝑜𝑢𝑟𝑠 “SHORTER”. This is our Ha.

 EXAMPLE 3
A Health Insurance claims that in 2011, the average
monthly premium paid for individual health coverage
was $183. Suppose you are suspicious that the average,
or mean, cost is actually higher. State the null and
alternative hypothesis you would use to test this.
In words,
The first statement must be
𝑡ℎ𝑒 𝑎𝑣𝑒𝑟𝑎𝑔𝑒 𝑚𝑜𝑛𝑡ℎ𝑙𝑦 𝑝𝑟𝑒𝑚𝑖𝑢𝑚 𝑓𝑜𝑟
Ho: ℎ𝑒𝑎𝑙𝑡ℎ 𝑐𝑜𝑣𝑒𝑟𝑎𝑔𝑒 𝑤𝑎𝑠 $183.
The NULL HYPOTHESIS since
the condition shows
Ha: ℎ𝑒𝑎𝑙𝑡ℎ 𝑐𝑜𝑣𝑒𝑟𝑎𝑔𝑒 𝑤𝑎𝑠 higher.
𝑡ℎ𝑒 𝑎𝑣𝑒𝑟𝑎𝑔𝑒 𝑚𝑜𝑛𝑡ℎ𝑙𝑦 𝑝𝑟𝑒𝑚𝑖𝑢𝑚 𝑓𝑜𝑟 EQUALITY.

In symbols, The second statement is the

ALTERNATIVE HYPOTHESIS
Ho: 𝜇 = $183 since the word “HIGHER”

Ha: 𝜇 > $183 indicates INEQUALITY.

 TYPES OF ERRORS
In testing the hypothesis, there only two decisions that
must be made:
1. REJECT the Null Hypothesis (Ho), and ACCEPT the
Alternative Hypothesis (Ha).
2. ACCEPT the Null Hypothesis (Ho), and REJECT the
Alternative Hypothesis (Ha).
In making decisions however, it is inevitable to commit
ERRORS. In statistics two possible type of errors can be
committed:
TYPE I ERROR is made if the NULL HYPOTHESIS (Ho) is
REJECTED when it is actually TRUE.

TYPE II ERROR is made if the NULL HYPOTHESIS (Ho) is

NOT REJECTED (ACCEPTED) when it is actually FALSE.
No matter which hypothesis represents the
claim, always begin the hypothesis test
assuming that the null hypothesis is true.

Actual Truth of H0
Decision H0 is true H0 is false

Do not reject H0 Correct Decision Type II Error

Reject H0 Type I Error Correct Decision

 IDENTIFYING ERRORS
EXAMPLE 1: In a court of law, the defendant is always
ASSUMED innocent of the crime until proven guilty.
a. State the Ho and Ha
b. Identify the Type I and Type II Errors in context
c. Describe the consequence for each type of error

Ho: The defendant is INNOCENT (NOT GULITY) of the crime.

Ha: The defendant is GUILTY of the crime.
For the Type I Error, Ho is REJECTED even though Ho is TRUE.
TYPE I ERROR:
The defendant is convicted GUILTY even though he is NOT.
For the Type II Error, Ho is ACCEPTED even though Ho is FALSE.
TYPE II ERROR:
The defendant is convicted INNOCENT even though HE IS NOT.
 The consequence for committing
TYPE I ERROR.
Punishing a person who is truly innocent
and putting them wrongly in jail.

The consequence for committing

TYPE II ERROR.
Criminals like the defendant gets away with
crimes and perhaps thinks he always can.
Later in life, this can lead to larger or
continued crimes.
 IDENTIFYING ERRORS
EXAMPLE 2: The USDA limit for salmonella contamination for
chicken is 20%. A meat inspector reports that the chicken
produced by a company exceeds the USDA limit. Use “p” and
not µ since
a. State the Ho and Ha data is in
b. Identify the Type I and Type II Errors in context percent

c. Describe the consequence for each type of error

Ho: The Salmonella contamination in chicken is 20% 𝒑 = 𝟐𝟎%
Ha: The Salmonella contamination in chicken exceeds 20% 𝒑 > 𝟐𝟎%
The second statement is the ALTERNATIVE HYPOTHESIS because the word
“EXCEEDS” indicate INEQUALITY which means GREATER THAN.
For the Type I Error, Ho is REJECTED and Ha is ACCEPTED even though Ho is TRUE.
TYPE I The meat inspector declared the Samonella content in chicken
ERROR: produced by the company exceeds 20% even though it is not
For the Type II Error, Ho is ACCEPTED even though Ho is FALSE.
TYPE II The meat inspector declared the Samonella content in chicken
ERROR: produced by the company is NORMAL (20%) when it is not.
 The consequence for committing
TYPE I ERROR.
Announcing that chicken exceeds the limit suggested by
the USDA even though it is not true may have big impact
to poultry raisers and vendors. Consumer would not
want to buy chicken and may cause in the downfall of
the poultry industry.
The consequence for committing
TYPE II ERROR.
Announcing that chicken are safe because the it is within
the limit set by the authority when actually it is NOT
would have a great impact to the CONSUMERS health.
In this case, many people would eventually get sick for
eating a contaminated chicken.
 PROBLEM SET # 4
A clean air standard requires that vehicle exhaust
emissions SHOULD NOT exceed specified limits for various
pollutants. Many states require cars be tested annually to be
sure they meet these standards. Suppose state regulators
double check a random sample of cars that a suspect repair
shop has certified as okay. They will revoke the shop’s
license if they find significant evidence that the shop is
certifying vehicles that do not meet standards.
a) What are the null and alternative hypotheses?
b) In this context, what is a Type I error?
c) In this context, what is a Type II error?
d) Which type of error would the shop’s owner consider more
serious? Why?
e) Which type of error might environmentalists consider more
serious? Why?
 Captain Ben flies small passenger jets. These
jets carry 50 passengers, plus their luggage. On a full
flight, these jets will perform properly as long as the
total weight of passengers’ checked baggage does
not exceed 5000 pounds. Ben is concerned that
passengers on a particular flight have brought
unusually heavy bags. He selects a random sample
of 10 passengers and weighs their checked baggage.
Based on the results from this sample, he must
decide whether it is safe to take off.

a) State the appropriate null and alternative hypotheses.

b) Describe the two types of errors that you might make.
Identify which is a Type I error and which is a Type II error.
c) Explain the consequences of each type of error. Which is
the most serious?
 A school administrator claims that students
whose first language is not English scored worse on
the verbal portion of the SAT exam than students
whose first language is English. The mean SAT
verbal score of students whose first language is
English is 515 with a population standard deviation
of 112, on the basis of data obtained from the College
Board. SAT verbal scores are normally distributed.

a) State the appropriate null and alternative hypotheses.

b) Describe the two types of errors that you might make.
Identify which is a Type I error and which is a Type II
error.
c) Explain the consequences of each type of error.
Which is the most serious?
 by:

LESSER A. LOC, MAEd

Project FREE - Paglaum
 HYPOTHESIS
TESTING
A hypothesis test is a process that uses
sample statistics to test a claim about the
value of a population parameter.
 Level of Significance
In a hypothesis test, the level of significance is your maximum allowable
probability of making a type I error. It is denoted by a, the lowercase Greek
letter alpha.

Hypothesis tests are

based on a.
The probability of making a type II error is denoted by b, the lowercase Greek
letter beta.

By setting the level of significance at a small value, you are saying that
you want the probability of rejecting a true null hypothesis to be small.

Commonly used levels of significance:

a = 0.10 a = 0.05 a = 0.01

 Statistical Tests
After stating the null and alternative hypotheses and
specifying the level of significance, a random sample is
taken from the population and sample statistics are
calculated.
The statistic that is compared with the parameter in
the null hypothesis is called the test statistic.
Population Test statistic Standardized test statistic
parameter

μ
x z (n ³ 30)
t (n < 30)
p p̂ z
s2 s2 X2
 How to write the
Null Hypothesis
The null hypothesis (Ho) must contain the
condition of equality and can be written in
the following ways:
𝐻𝑜: 𝜇 = 𝑘 𝐻𝑜: 𝜇 ≤ 𝑘 𝐻𝑜: 𝜇 ≥ 𝑘

Word clues: equal to, at least (greater than

or equal to), at most (less than or equal to)

Note: k could be any value

 How to write the
Alternative Hypothesis
The alternative hypothesis (Ha) is a statement that
challenges the null hypothesis. Hence, the null
hypothesis must contain the condition of
inequality. For the mean, the alternative hypothesis
can be stated in only one of these three possible
forms:

𝐻𝑎: 𝜇 ≠ 𝑘 𝐻𝑎: 𝜇 > 𝑘 𝐻𝑎: 𝜇 < 𝑘

Word clues: different, greater than, less than

Note: k could be any value

 Types of Hypothesis Tests
There are two types of hypothesis tests
– one-tailed or two-tailed test.

The one-tailed test could either be left-

tailed or right-tailed test.

The type of test depends on the region

of the sampling distribution that favors a
rejection of H0. This region is indicated
by the alternative hypothesis.
 One-Tailed Test
1. If the alternative hypothesis contains the less-than (<) or
greater than inequality symbol (>), the hypothesis test is a
one-tailed test.
Since the alternative hypothesis (Ha) uses
H 0: µ ³ k the LESS THAN inequality symbol (<),
Example:
H a: µ < k then the test is left-tailed.

P is the area to
the left of the
test statistic.

z
-3 -2 -1 0 1 2 3
Test
statistic
 Since the alternative hypothesis (Ha) uses
H 0: µ £ k the GREATER THAN inequality symbol
EXAMPLE 2: (>), then the test is RIGHT-TAILED.
H a: µ > k

P is the area to
the right of the
test statistic.

z
-3 -2 -1 0 1 2 3
Test
statistic
 Two-tailed Test
3. If the alternative hypothesis contains the not-equal-to
symbol (¹), the hypothesis test is a two-tailed test. In a
1
two-tailed test, each tail has an area of 2 P.
H0 : µ = k
Ha : µ ¹ k
P is twice the
P is twice the area to the right
area to the left of the positive
of the negative test statistic.
test statistic.

z
-3 -2 -1 0 1 2 3
Test Test
statistic statistic
 Identifying Types of Tests
Example:
For each claim, state H0 and Ha. Then determine whether the
hypothesis test is a left-tailed, right-tailed, or two-tailed test.

a.) A cigarette manufacturer claims that less than one-

eighth of the US adult population smokes cigarettes.
H0: p ³ 0.125
Ha: p < 0.125 (Claim) One-Tailed Test (Left-tailed)

b.) A local telephone company claims that the average

length of a phone call is 8 minutes.
H0: μ = 8 (Claim)
Ha : μ ¹ 8 Two-tailed test
 STEPS ON HYPOTHESIS TESTING
State the alternative and null hypothesis

Select the level of significance

Determine the critical value and the rejection region

State the decision rule

Compute the test statistic

Make a decision, whether to reject or not to reject the null hypothesis.

REJECT Ho, If COMPUTED VALUE is GREATER THAN the TABULAR (CRITICAL) VALUE

ACCEPT Ho, If COMPUTED VALUE is LESS THAN the TABULAR (CRITICAL) VALUE
Hypothesis Test for
LARGE SAMPLE
SIZE
(𝒏 ≥ 𝟑𝟎)
 ONE SAMPLE Z-TEST
The z – test according to Pagano(2010) is
appropriate tool for measuring the significant
difference between the two groups of more than 30
respondents or sample.

ASSUMPTIONS:
The central limit theorem is applied if the sample is
large (n > 30). In this case you will use the normal
distribution (z-distribution).
In applying the central limit theorem, you can use the
sample standard deviation s as an estimate of the population
standard deviation , whenever is not known and the sample
size is large.
 FORMULA FOR ONE-SAMPLE Z-TEST
𝑥̅ − 𝜇
𝑧=𝜎
' 𝑛

CRITICAL VALUES FOR Z

Level of Significance (∞)
Test Type
0.01 0.025 0.05 0.10
One-Tailed Test
±2.33 ±1.96 ±1.645 ±1.28
(Left-Tailed or Right-Tailed)

Two-Tailed Test ±2.575 ±2.33 ±1.96 ±1.645

NOW LET’S DO
HYPOTHESIS
TESTING
 EXAMPLE#1 GIVEN: 𝝁 = 𝑷𝟏𝟐𝟐𝟐. 𝟎𝟎
According to the national
survey, young men ages 8-17 𝒏 = 𝟓𝟎 𝒎𝒆𝒏
years old spend an average
- = 𝑷𝟏, 𝟏𝟔𝟖. 𝟓𝟎
𝒙
of P1222.00 per shopping
trip to a local mall. A 𝝈 = 𝑷𝟏𝟖𝟓. 𝟎𝟎
manager of a local mall
surveyed 50 young men and ∝ = 𝟎. 𝟎𝟓
found out that the average STEP I:
amount spent per trip was
P1,168.50 with a standard State Ho and Ha:
deviation of P185.00. With Ho: 𝝁 = 𝑷𝟏𝟐𝟐𝟐. 𝟎𝟎
Alpha = 0.05, can it be
Ha: 𝝁 ≠ 𝑷𝟏𝟐𝟐𝟐. 𝟎𝟎
concluded that 8-17 years
old spend a different
(two-tailed test)
amount than the national
average at the local mall?
STEP II: Select Level of Significance 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 =
−𝟓𝟑. 𝟓
𝟐𝟔. 𝟏𝟔𝟐𝟗𝟓𝟏
∝ = 𝟎. 𝟎𝟓
STEP III: Determine Critical Region 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = −𝟐. 𝟎𝟒
𝑺𝒊𝒏𝒄𝒆 ∝ = 𝟎. 𝟎𝟓, 𝐚𝐧𝐝 𝐭𝐰𝐨 − 𝐭𝐚𝐢𝐥𝐞𝐝. STEP V: Make a DECISION
𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 = ±𝟏. 𝟗𝟔 (Data taken from the table on slide #15) 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = −𝟐. 𝟎𝟒
STEP IV: State the DECISION RULE 𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 = ±𝟏. 𝟗𝟔
𝑹𝑬𝑱𝑬𝑪𝑻 𝑯𝒐 𝒊𝒇 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 > 𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 Since, 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 > 𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳

𝑨𝑪𝑪𝑬𝑷𝑻 𝑯𝒐 𝒊𝒇 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 < 𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 Then, 𝒕𝒉𝒆 𝑫𝑬𝑪𝑰𝑺𝑰𝑶𝑵 𝒊𝒔 𝒕𝒐

𝑹𝑬𝑱𝑬𝑪𝑻 𝒕𝒉𝒆 𝑵𝑼𝑳𝑳 𝑯𝒀𝑷𝑶𝑻𝑯𝑬𝑺𝑰𝑺
STEP V: Solve for the TEST STATISTICS
(𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 ) STEP VI: CONCLUSION
=−𝝁
𝒙 Therefore, it can be
𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 =𝝈
@ 𝒏 CONCLUDED that 8-17 years
old spend a different amount
𝟏, 𝟏𝟔𝟖. 𝟓𝟎 − 𝟏, 𝟐𝟐𝟐
𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 =
𝟏𝟖𝟓@
than the national average at
𝟓𝟎 the local mall.
 EXAMPLE 2 GIVEN: 𝝁 = 𝟏𝟎. 𝟔 𝒈/𝒎𝑳
The hemoglobin levels of
children below 6 years old in a 𝒏 = 𝟓𝟎 𝒄𝒉𝒊𝒍𝒅𝒓𝒆𝒏
place where there is high level of
lead tend to be high also. - = 𝟏𝟏. 𝟏 𝒈/𝒎𝑳
𝒙
According to Medical experts the
tolerable hemoglobin level is 𝝈 = 𝟓. 𝟎𝟏 𝒈/𝒎𝑳
10.6 g/100mL. In a certain city a
test was conducted to 50 ∝ = 𝟎. 𝟎𝟏
children selected at random and STEP I:
it was found out the average
hemoglobin level of the 50 State Ho and Ha:
children is 11.1 g/100ml and its
standard deviation, s = 5.01 Ho: 𝝁 = 𝟏𝟎. 𝟔 𝒈/𝒎𝑳 because of
the word
g/100ml. Is there a reason to Ha: 𝝁 > 𝟏𝟎. 𝟔 𝒈/𝒎𝑳 higher
believe that the hemoglobin of
the children in this city is higher (one-tailed test)
than normal? Test at ∞ = 0.01.
STEP II: Select Level of Significance 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 =
𝟎. 𝟓
𝟎. 𝟕𝟎𝟖𝟓𝟐𝟏
∝ = 𝟎. 𝟎𝟏
STEP III: Determine Critical Region 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = 𝟎. 𝟕𝟏
𝑺𝒊𝒏𝒄𝒆 ∝= 𝟎. 𝟎𝟏 𝒂𝒏𝒅 𝒐𝒏𝒆 − 𝒕𝒂𝒊𝒍𝒆𝒅 STEP V: Make a DECISION
𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 = ±𝟐. 𝟑𝟑 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = 𝟎. 𝟕𝟏
STEP IV: State the DECISION RULE 𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 = ±𝟐. 𝟑𝟑
𝑹𝑬𝑱𝑬𝑪𝑻 𝑯𝒐 𝒊𝒇 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 > 𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 Since, 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 < 𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳
𝑨𝑪𝑪𝑬𝑷𝑻 𝑯𝒐 𝒊𝒇 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 < 𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 Then, 𝒕𝒉𝒆 𝑫𝑬𝑪𝑰𝑺𝑰𝑶𝑵 𝒊𝒔 𝒕𝒐
𝑨𝑪𝑪𝑬𝑷𝑻 𝒕𝒉𝒆 𝑵𝑼𝑳𝑳 𝑯𝒀𝑷𝑶𝑻𝑯𝑬𝑺𝑰𝑺
STEP IV: Solvefor the TEST
Ho: 𝝁 = 𝟏𝟎. 𝟔 𝒈/𝒎𝑳
STATISTICS (𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 )
=−𝝁
𝒙 STEP VI: CONCLUSION
𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = 𝝈
@ 𝒏
Therefore, there is NOT enough
REASON to believe that the
𝟏𝟏. 𝟏 − 𝟏𝟎. 𝟔
𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = Hemoglobin level of children in the
𝟓. 𝟎𝟏@
𝟓𝟎 city is higher than normal.
 EXAMPLE 3
GIVEN: 𝝁 = 𝟎. 𝟔𝟏 𝒔𝒆𝒄
A test on car braking reaction
times for men between 18 𝝈 = 𝟎. 𝟏𝟐𝟑 𝒔𝒆𝒄
and 30 years old has
produced a mean and a
𝒏 = 𝟒𝟎 𝒅𝒓𝒊𝒗𝒆𝒓𝒔
standard deviation of 0.61 sec - = 𝟎. 𝟓𝟖𝟕 𝒔𝒆𝒄
𝒙
and 0.123 sec, respectively.
When 40 male drivers of this ∝ = 𝟎. 𝟏𝟎
group were randomly
selected tested for their STEP I:
breaking reaction times, a
mean of 0.587 second came State Ho and Ha:
out. At the 0.10 level of
Ho: 𝝁 = 𝟎. 𝟔𝟏 𝒔𝒆𝒄
significance, test claim of the
driving instructor that his Ha: 𝝁 < 𝟎. 𝟔𝟏 𝒔𝒆𝒄
graduates had faster reaction
times. (one-tailed test)
STEP II: Select Level of Significance 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 =
−𝟎. 𝟎𝟐𝟑
𝟎. 𝟎𝟏𝟗𝟒𝟒𝟖
∝ = 𝟎. 𝟏𝟎
STEP III: Determine Critical Region 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = −𝟏. 𝟏𝟖
𝑺𝒊𝒏𝒄𝒆 ∝= 𝟎. 𝟏𝟎 𝒂𝒏𝒅 𝒐𝒏𝒆 − 𝒕𝒂𝒊𝒍𝒆𝒅 STEP V: Make a DECISION
𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 = ±𝟏. 𝟐𝟖 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = −𝟏. 𝟏𝟖
STEP IV: State the DECISION RULE 𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 = ±𝟏. 𝟐𝟖
𝑹𝑬𝑱𝑬𝑪𝑻 𝑯𝒐 𝒊𝒇 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 > 𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 Since, 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 < 𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳

𝑨𝑪𝑪𝑬𝑷𝑻 𝑯𝒐 𝒊𝒇 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 < 𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 Then, 𝒕𝒉𝒆 𝑫𝑬𝑪𝑰𝑺𝑰𝑶𝑵 𝒊𝒔 𝒕𝒐

𝑨𝑪𝑪𝑬𝑷𝑻 𝒕𝒉𝒆 𝑵𝑼𝑳𝑳 𝑯𝒀𝑷𝑶𝑻𝑯𝑬𝑺𝑰𝑺
STEP IV: Solve
for the TEST
Ho: 𝝁 = 𝟎. 𝟔𝟏 𝒔𝒆𝒄
STATISTICS (𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 )
=−𝝁
𝒙
STEP VI: CONCLUSION
𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 =𝝈
@ 𝒏
Therefore, there is NOT enough
evidence to support the claim of the
𝟎. 𝟓𝟖𝟕 − 𝟎. 𝟔𝟏 driving instructor that his graduates
𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 =
𝟎. 𝟏𝟐𝟑@ has faster reaction time.
𝟒𝟎
 TWO SAMPLE Z-TEST
Otherwise known as z-test for two independent
samples. In this type of z-test, we compare the
means between TWO GROUPS (SAMPLES).

Formula: 𝑥! − 𝑥"
𝑧=
𝜎! "
𝜎" "
𝑛! + 𝑛"
It is said that the solvent known
This problem is an example of a
as benzene is found in cigarettes
TWO-SAMPLE z-test because we are
and cigars. A random sample of
going to compare two GROUPS:
50 cigars were found to have
CIGARS and CIGARETTES.
benzene concentration of
150 𝑔/𝜇𝑔 with a standard GIVEN
deviation of 10 𝑔/𝜇𝑔 while a GROUPS SAMPLE SAMPLE MEAN STD
random sample of 40 cigarettes (n) (!
𝒙) DEVIATION
(𝝈)
were found to have a benzene CIGARS 𝑛! = 50 𝑥̅! = 150𝑔/𝜇𝑔 𝜎! = 10𝑔/𝜇𝑔
concentration of 147 𝑔/𝜇𝑔 with
a standard deviation of 8 𝑔/𝜇𝑔 . CIGARETTES 𝑛" = 40 𝑥̅" = 147𝑔/𝜇𝑔 𝜎" = 8 𝑔/𝜇𝑔

Is it statistically correct to
conclude that cigars in this Step I: State Ho and Ha
particular hypothetical Ho: 𝜇_`abcd = 𝜇_`abceffed
experiment have higher benzene
concentration than cigarettes?
Ha: 𝜇_`abcd > 𝜇_`abceffed
Test this hypothesis at 5% level ONE-TAILED TEST
of significance.
 𝟑
STEP II: Select Level of Significance 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 =
𝟏𝟎𝟎 𝟔𝟒
+
∝ = 𝟎. 𝟎𝟓 𝟓𝟎 𝟒𝟎
𝟑
STEP III: Determine Critical Region 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 =
𝟐 + 𝟏. 𝟔
𝑺𝒊𝒏𝒄𝒆 ∝= 𝟎. 𝟎𝟓 𝒂𝒏𝒅 𝒐𝒏𝒆 − 𝒕𝒂𝒊𝒍𝒆𝒅 𝟑 𝟑
𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = =
𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 = ±𝟏. 𝟔𝟒𝟓 𝟑. 𝟔 𝟏. 𝟖𝟗𝟕𝟑𝟔𝟕
𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = 𝟏. 𝟓𝟖
STEP IV: State the DECISION RULE
STEP V: Make a DECISION
𝑹𝑬𝑱𝑬𝑪𝑻 𝑯𝒐 𝒊𝒇 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 > 𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳
𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = 𝟏. 𝟓𝟖
𝑨𝑪𝑪𝑬𝑷𝑻 𝑯𝒐 𝒊𝒇 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 < 𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳
𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 = ±𝟏. 𝟔𝟒𝟓
STEP IV: Solve for the
TEST STATISTICS (𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 ) Since, 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 < 𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳
𝑥* − 𝑥+ Then, 𝒕𝒉𝒆 𝑫𝑬𝑪𝑰𝑺𝑰𝑶𝑵 𝒊𝒔 𝒕𝒐
𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 =
𝜎* + 𝜎+ + 𝐀𝐂𝐂𝐄𝐏𝐓 𝒕𝒉𝒆 𝑵𝑼𝑳𝑳 𝑯𝒀𝑷𝑶𝑻𝑯𝑬𝑺𝑰𝑺
+
𝑛* 𝑛+
STEP VI: CONCLUSION
𝟏𝟓𝟎 − 𝟏𝟒𝟕 Therefore, it is NOT CORRECT to conclude that cigars
𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = have higher BENZENE CONCENTRATION than
𝟏𝟎𝟐 𝟖𝟐
+ CIGARETTES. It only implies that benzene concentration
𝟓𝟎 𝟒𝟎
of cigars and cigarettes are almost the same.
In two countries C1 and C2,
the heights of women were This problem is an example of a
TWO-SAMPLE z-test because we are
studied. From country C1, going to compare two GROUPS:
120 women were measured COUNTRY 1 & COUNTRY 2.
on their heights and the
GIVEN
mean height is 62.7 inches
with a standard deviation of GROUPS SAMPLE (n)
SAMPLE MEAN
(!
𝒙)
STD
DEVIATION
2.50, while in other country (𝝈)
C1 𝑛 = 120 𝑥̅ = 62.7 𝑖𝑛 𝜎 = 2.50 𝑖𝑛
C2, the mean height is 61.8
! ! !

C2 𝑛 = 150
" 𝑥̅ = 61.8 𝑖𝑛
" 𝜎 = 2.62 𝑖𝑛
"

with a standard deviation of

2.62 inches on 150 women. Step I: State Ho and Ha
Is there a significant reason
Ho: 𝜇_k = 𝜇_l
to believe that the heights
of women in C1 and C2 Ha: 𝜇_k ≠ 𝜇_l
significantly differ? Use 0.01 TWO-TAILED TEST
level of significance.
 𝟎. 𝟗
STEP II: Select Level of Significance 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 =
𝟔. 𝟐𝟓 𝟔. 𝟖𝟔𝟒𝟒
∝ = 𝟎. 𝟎𝟏 𝟏𝟐𝟎
+
𝟏𝟓𝟎
𝟎. 𝟗
STEP III: Determine Critical Region 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 =
𝟎. 𝟎𝟓𝟐𝟎𝟖𝟑 + 𝟎. 𝟎𝟒𝟓𝟕𝟔𝟑
𝑺𝒊𝒏𝒄𝒆 ∝= 𝟎. 𝟎𝟏 𝒂𝒏𝒅 𝒕𝒘𝒐 − 𝒕𝒂𝒊𝒍𝒆𝒅 𝟎. 𝟗 𝟎. 𝟗
𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = =
𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 = ±𝟐. 𝟓𝟕𝟓 𝟎. 𝟎𝟗𝟕𝟖𝟒𝟔 𝟎. 𝟑𝟏𝟐𝟖𝟎𝟑
𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = 𝟐. 𝟖𝟖
STEP IV: State the DECISION RULE
STEP V: Make a DECISION
𝑹𝑬𝑱𝑬𝑪𝑻 𝑯𝒐 𝒊𝒇 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 > 𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳
𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = 𝟐. 𝟖𝟖
𝑨𝑪𝑪𝑬𝑷𝑻 𝑯𝒐 𝒊𝒇 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 < 𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳
𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 = ±𝟐. 𝟓𝟕𝟓
STEP IV: Solve for the
TEST STATISTICS (𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 ) Since, 𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 > 𝒁𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳
𝑥* − 𝑥+ Then, 𝒕𝒉𝒆 𝑫𝑬𝑪𝑰𝑺𝑰𝑶𝑵 𝒊𝒔 𝒕𝒐
𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 =
𝜎* + 𝜎+ + REJECT 𝒕𝒉𝒆 𝑵𝑼𝑳𝑳 𝑯𝒀𝑷𝑶𝑻𝑯𝑬𝑺𝑰𝑺
+
𝑛* 𝑛+
STEP VI: CONCLUSION
𝟔𝟐. 𝟕 − 𝟔𝟏. 𝟖 Therefore, there is a significant reason to
𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 =
𝟐. 𝟓𝟎𝟐 𝟐. 𝟔𝟐𝟐 believe that the heights of women in C1
𝟏𝟐𝟎 + 𝟏𝟓𝟎
and C2 significantly differ
 Problem Set #5
1. The required amount of softdrinks in a
bottling company should be 500 mL. A
quality control expert of the company took a
35 randomly selected bottles from a batch
and measured the amount of liquids in each
bottle sample. The amounts in the sample
had a mean of 503 mL and a standard
deviation of 5 mL. Is it correct to conclude
that the mean amount of bottles is different
than the required amount with alpha = 0.10?
 Problem Set #5
2. According to the U.S. Department of
Education, full-time graduates receive an
average salary of $12,800. The dean of
graduate studies at a large university claims
that their graduates earn more than the
national average. He surveys 46 randomly
selected students and finds their average
salary is $13,445 with a standard deviation
of $1800. With alpha = 0.05, is there enough
evidence to support the Dean’s claim?
 Problem Set #5
3. A high school math teacher claims that
students in her class will score higher on math
portion of the NAT than students in a
colleague’s math class. The mean NAT Math
score for 49 student in her class is 22.1 and the
standard deviation is 4.8. On the other hand,
the man NAT Math score for 44 of the
Colleague’s students in 19.8 and the standard
deviation is 5.4. At alpha = 0.10, is there
enough evidence to support the claim of the
teacher that her students will score better on
math portion of the NAT than students in her
colleague’s math class?
 Problem Set #5
4. The amount of a certain trace element in
blood is known to vary with a standard
deviation of 14.1 ppm (parts per million) for
male blood donors and 9.5 ppm for female
donors. Random sample of 75 male and 50
female donors yield concentration means
of 28 and 33 ppm, respectively. Is it correct
to conclude that the population means of
concentrations of the element significantly
differ for male and female donors. Test at
alpha = 0.01.
Hypothesis Test for
SMALL SAMPLE
SIZE
(𝒏 < 𝟑𝟎)
The t-test is used when sample
size is less than 30 together with
the assumption that the
population where the sample
comes from is normally
distributed.
ONE-SAMPLE T-TEST

𝑥̅ − 𝜇
𝑡=𝑠
' 𝑛
 EXAMPLE 1 GIVEN:
Medical literature tells us that
𝝁 = 𝟕𝟐𝟓𝟎/𝒎𝒎𝟑
our body is mainly composed
of red and white corpuscles 𝒏 = 𝟏𝟓
and a normal human body + = 𝟒𝟖𝟓𝟎/𝒎𝒎𝟑
𝒙
must average 7250/mm3 of 𝒔 = 𝟐𝟓𝟎𝟎/𝒎𝒎𝟑
white blood cell counts. If a
sample of 15 individuals STATE Ho and Ha:
chosen at random from a
Ho: 𝝁 = 𝟕𝟐𝟓𝟎/𝒎𝒎𝟑
certain place have an average
of 4850/mm3 with a standard Ha: 𝝁 < 𝟕𝟐𝟓𝟎/𝒎𝒎𝟑
deviation of 2500/mm3, would
you say that the people in that (One-Tailed Test)
place have low white blood cell
count? Test at 5% level of
significance.
 Identify Significance Level (∞)
and Degrees of Freedom (df)

∝= 𝟎. 𝟎𝟓
d𝒇 = 𝒏 − 𝟏
d𝒇 = 𝟏𝟓 − 𝟏
d𝒇 = 𝟏𝟒
Determine critical value for
(𝒕𝒄𝒓𝒊𝒕𝒊𝒄𝒂𝒍 )
(𝐓𝐡𝐞 𝒕 − 𝒅𝒊𝒔𝒕𝒓𝒊𝒃𝒖𝒕𝒊𝒐𝒏 𝒘𝒊𝒍𝒍 𝒃𝒆 𝒖𝒔𝒆𝒅)

𝑺𝒊𝒏𝒄𝒆:
𝒐𝒏𝒆 − 𝒕𝒂𝒊𝒍𝒆𝒅
∝= 𝟎. 𝟎𝟓
𝒅𝒇 = 𝟏𝟒
𝒕𝒄𝒓𝒊𝒕𝒊𝒄𝒂𝒍 = 𝟏. 𝟕𝟔𝟏
DECISION RULE:
𝑹𝑬𝑱𝑬𝑪𝑻 𝑯𝒐 𝒊𝒇 𝒕𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 > 𝒕𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳
𝑨𝑪𝑪𝑬𝑷𝑻 𝑯𝒐 𝒊𝒇 𝒕𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 < 𝒕𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳

Compute for the test statistic

(𝒕𝒄𝒐𝒎𝒑𝒖𝒕𝒆𝒅)

𝑥̅ − 𝜇 4850 − 7250 −2400

𝑡=𝑠 = = = −3.72
K 𝑛 2500K 645.497224
15
MAKE DECISION

𝑡wxyz{|}~ = −3.72 𝑡w•€|€w•‚ = 1.761

𝑺𝒊𝒏𝒄𝒆 𝒕𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 > 𝒛𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 , then REJECT Ho.
CONCLUSION
It is significantly correct to say that people in that place
have low white blood cell count than normal.
 EXAMPLE 2 GIVEN:
The manager rental car
agency claims that the 𝝁 = 𝟖𝟎𝟎𝟎 𝒎𝒊𝒍𝒆𝒔
average mileage of cars 𝒏=𝟓
rented is less than 8000. A + = 𝟕𝟕𝟐𝟑 𝒎𝒊𝒍𝒆𝒔
𝒙
sample of five 𝒔 = 𝟓𝟎𝟎 𝒎𝒊𝒍𝒆𝒔
automobiles has an STATE Ho and Ha:
average mileage of 7723, Ho: 𝝁 = 𝟖𝟎𝟎𝟎 𝒎𝒊𝒍𝒆𝒔
with a standard deviation Ha: 𝝁 < 𝟖𝟎𝟎𝟎 𝒎𝒊𝒍𝒆𝒔
of 500 miles. At 0.01 level
of significance, is there (One-Tailed Test)
enough evidence to
support the managers’
claim?
 Identify Significance Level (∞)
and Degrees of Freedom (df)

∝= 𝟎. 𝟎𝟏
d𝒇 = 𝒏 − 𝟏
d𝒇 = 𝟓 − 𝟏
d𝒇 = 𝟒
Determine critical value for
(𝒕𝒄𝒓𝒊𝒕𝒊𝒄𝒂𝒍 )
(𝐓𝐡𝐞 𝒕 − 𝒅𝒊𝒔𝒕𝒓𝒊𝒃𝒖𝒕𝒊𝒐𝒏 𝒘𝒊𝒍𝒍 𝒃𝒆 𝒖𝒔𝒆𝒅)

𝑺𝒊𝒏𝒄𝒆:
𝒐𝒏𝒆 − 𝒕𝒂𝒊𝒍𝒆𝒅
∝= 𝟎. 𝟎𝟏
𝒅𝒇 = 𝟒
𝒕𝒄𝒓𝒊𝒕𝒊𝒄𝒂𝒍 = 𝟑. 𝟕𝟒𝟕
DECISION RULE:
𝑹𝑬𝑱𝑬𝑪𝑻 𝑯𝒐 𝒊𝒇 𝒕𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 > 𝒕𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳
𝑨𝑪𝑪𝑬𝑷𝑻 𝑯𝒐 𝒊𝒇 𝒕𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 < 𝒕𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳

Compute for the test statistic

(𝒕𝒄𝒐𝒎𝒑𝒖𝒕𝒆𝒅)

𝑥̅ − 𝜇 7723 − 8000 −277

𝑡=𝑠 = = = −1.24
K 𝑛 500K 223.606798
5
MAKE DECISION

𝑡-./01234 = −1.24 𝑡-5626-78 = 3.747

𝑺𝒊𝒏𝒄𝒆 𝒕𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 < 𝒛𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 , then ACCEPT Ho.

There is not enough evidence to support

CONCLUSION the claim of the manager that the average
mileage of rented cars is less than 8000
miles.
Two-Sample t-test (Independent Sample t-test)
DIFFERENCES OF TWO SAMPLE MEANS WITH EQUAL VARIANCES
When sample sizes are small and the variances are statistically equal, the
vriances will then be pooled and the degrees of freedom is computed as . The
test statistic to be used will be the following:
𝑥j − 𝑥k WHERE:
𝑡= 𝑛j − 1 𝑠jk + (𝑛k − 1)𝑠kk
𝑠l
k
𝑠l
k 𝑠l k =
+ 𝑑𝑓
𝑛j 𝑛k
DIFFERENCES OF TWO SAMPLE MEANS WITH UNEQUAL VARIANCES
When sample sizes are small and the variances of the populations where the
samples come from cannot be ascertained to be statistically equal, then the
variances will not be pooled and the degrees of freedom is computed by the
following formula:
𝑠jk 𝑠k k k
𝑥j − 𝑥k 𝑛j + 𝑛k
𝑡= 𝑑𝑓 =
𝑠j k 𝑠k k k k
𝑠j 𝑠k k k
+
𝑛j 𝑛k 𝑛j 𝑛k
+
𝑛j − 1 𝑛k − 1
 EXAMPLE 1
Is there a significant difference between the lifetime of batteries made by
Duracell and Everyday given the data below. Test the hypothesis at 0.05
level of significance. Assume that variances are equal.

DURACELL EVERYDAY
n 8 10
Mean 41 45
Std Deviation 18 20
GIVEN:
STATE Ho and Ha:
𝒏𝟏 = 𝟖 𝒏𝟐 = 𝟏𝟎
Ho: 𝝁𝑫𝑼𝑹𝑨𝑪𝑬𝑳𝑳 = 𝝁𝑬𝑽𝑬𝑹𝒀𝑫𝑨𝒀
𝒙 +𝟐 = 𝟒𝟓
+𝟏 = 𝟒𝟏 𝒙 Ha: 𝝁𝑫𝑼𝑹𝑨𝑪𝑬𝑳𝑳 ≠ 𝝁𝑬𝑽𝑬𝑹𝒀𝑫𝑨𝒀
𝒔𝟏 = 𝟏𝟖 𝒔𝟐 = 𝟐𝟎 (Two-Tailed Test)
 Identify Significance Level (∞)
and Degrees of Freedom (df)

∝= 𝟎. 𝟎𝟓
d𝒇 = 𝒏𝟏 + 𝒏𝟐 − 𝟐
d𝒇 = 𝟖 + 𝟏𝟎 − 𝟐
d𝒇 = 𝟏𝟔
Determine critical value for
(𝒕𝒄𝒓𝒊𝒕𝒊𝒄𝒂𝒍 )
(𝐓𝐡𝐞 𝒕 − 𝒅𝒊𝒔𝒕𝒓𝒊𝒃𝒖𝒕𝒊𝒐𝒏 𝒘𝒊𝒍𝒍 𝒃𝒆 𝒖𝒔𝒆𝒅)

𝑺𝒊𝒏𝒄𝒆:
𝒕𝒘𝒐 − 𝒕𝒂𝒊𝒍𝒆𝒅
∝= 𝟎. 𝟎𝟓
𝒅𝒇 = 𝟏𝟔
𝒕𝒄𝒓𝒊𝒕𝒊𝒄𝒂𝒍 = 𝟐. 𝟏𝟐𝟎
 𝟐
DECISION RULE: FIRST SOLVE FOR 𝒔𝒑
𝑹𝑬𝑱𝑬𝑪𝑻 𝑯𝒐 𝒊𝒇 𝒕𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 > 𝒕𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 𝑛= − 1 𝑠= > + (𝑛> − 1)𝑠> >
𝑠? > =
𝑨𝑪𝑪𝑬𝑷𝑻 𝑯𝒐 𝒊𝒇 𝒕𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 < 𝒕𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 𝑑𝑓

Compute for the test statistic 8 − 1 (18)> +(10 − 1)(20)>

𝑠? > =
(𝒕𝒄𝒐𝒎𝒑𝒖𝒕𝒆𝒅) for EQUAL VARIANCE 16
𝑥= − 𝑥> 7 (324) + (9)(400)
𝑡56789:;< = 𝑠? > =
> > 16
𝑠? 𝑠? 5868
+ 𝑠? > = = 𝟑𝟔𝟔. 𝟕𝟓
𝑛= 𝑛> 16
41 − 45 −4 −4 −4
𝑡23456789 = = = = = −𝟎. 𝟒𝟒
366.75 366.75 45.84375 + 36.675 82.51875 9.083983
+
8 10

MAKE DECISION 𝑡z{|l}~•€ = −0.44 𝑡z•‚~‚zƒ„ = 2.120

𝑺𝒊𝒏𝒄𝒆 𝒕𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 < 𝒛𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 , then ACCEPT Ho.
CONCLUSION
There is no significant difference between the lifetime of batteries made by
Duracell and Everyday.
 EXAMPLE 2
Many people take gingko supplements advertised to improve memory. Are these over-
the-counter supplements effective? In a study, elderly adults were assigned to the
treatment group or control group. The 10 participants who were assigned to the
treatment group took 40 mg of gingko 3 times a day for 6 weeks. The 15 participants
assigned to the control group took did not take any supplement for 6 weeks. At the
end of 6 weeks, the Wechsler Memory Scale was administered. Higher scores
indicate better memory function. Summary values are given in the following table:
Based on the results, is there enough
n Mean Std Deviation evidence to support the claim that taking 40
TREATMENT 10 5.7 0.6 mg Gingko 3 times a day for 6 weeks is
CONTROL 15 5.5 0.5 effective in increasing the mean
performance on the Wechsler Memory Scale
at 0.05 level of significance? ASSUME THAT
GIVEN: VARIANCES ARE NOT EQUAL

𝒏𝟏 = 𝟏𝟎 𝒏𝟐 = 𝟏𝟓 STATE Ho and Ha:

Ho: 𝝁𝑻𝑹𝑬𝑨𝑻𝑴𝑬𝑵𝑻 = 𝝁𝑪𝑶𝑵𝑻𝑹𝑶𝑳
𝒙 +𝟐 = 𝟓. 𝟓
+𝟏 = 𝟓. 𝟕 𝒙 Ha: 𝝁𝑻𝑹𝑬𝑨𝑻𝑴𝑬𝑵𝑻 > 𝝁𝑪𝑶𝑵𝑻𝑹𝑶𝑳
𝒔𝟏 = 𝟎. 𝟔 𝒔𝟐 = 𝟎. 𝟓 (One-Tailed Test)
Identify Significance Level (∞) 𝑻𝒉𝒆𝒏 𝒔𝒐𝒍𝒗𝒆 𝒇𝒐𝒓 𝒅𝒇
and Degrees of Freedom (df)
0.036 + 0.0167 >
𝑑𝑓 =
∝= 𝟎. 𝟎𝟓 0.036 >
10 − 1
+
0.0167 >
15 − 1
𝑻𝒉𝒆 𝒅𝒆𝒈𝒓𝒆𝒆𝒔 𝒐𝒇 𝒇𝒓𝒆𝒆𝒅𝒐𝒎 𝒇𝒐𝒓
𝑼𝑵𝑬𝑸𝑼𝑨𝑳 𝑽𝑨𝑹𝑰𝑨𝑵𝑪𝑬𝑺 𝒉𝒂𝒗𝒆
𝑫𝑰𝑭𝑭𝑬𝑹𝑬𝑵𝑻 𝑭𝑶𝑹𝑴𝑼𝑳𝑨: 0.0527 >
𝑑𝑓 =
> > > 0.036 > 0.0167 >
𝑠= 𝑠 +
+ > 9 14
𝑛= 𝑛>
𝑑𝑓 = >
𝑠= > 𝑠> >
> 0.00277729
𝑑𝑓 =
𝑛= 𝑛 0.000144 + 0.000020
+ >
𝑛= − 1 𝑛> − 1
0.00277729
(𝒔𝟏 )𝟐
(𝒔𝟐 )𝟐 𝑑𝑓 =
𝑭𝒊𝒓𝒔𝒕 𝒔𝒐𝒍𝒗𝒆 𝒇𝒐𝒓 𝒂𝒏𝒅 0.000144 + 0.000020
𝒏𝟏 𝒏𝟐
(𝒔𝟏 )𝟐 (𝟎. 𝟔)𝟐 0.00277729
= = 𝟎. 𝟎𝟑𝟔 𝑑𝑓 =
𝒏𝟏 𝟏𝟎 0.000164
𝑑𝑓 = 16.94
(𝒔𝟐 )𝟐 (𝟎. 𝟓)𝟐
= = 𝟎. 𝟎𝟏𝟔𝟕
𝒏𝟐 𝟏𝟓 𝒅𝒇 ≈ 𝟏𝟕
Determine critical value for
(𝒕𝒄𝒓𝒊𝒕𝒊𝒄𝒂𝒍 )
(𝐓𝐡𝐞 𝒕 − 𝒅𝒊𝒔𝒕𝒓𝒊𝒃𝒖𝒕𝒊𝒐𝒏 𝒘𝒊𝒍𝒍 𝒃𝒆 𝒖𝒔𝒆𝒅)

𝑺𝒊𝒏𝒄𝒆:
𝒐𝒏𝒆 − 𝒕𝒂𝒊𝒍𝒆𝒅
∝= 𝟎. 𝟎𝟓
𝒅𝒇 = 𝟏𝟕
𝒕𝒄𝒓𝒊𝒕𝒊𝒄𝒂𝒍 = 𝟏. 𝟕𝟒𝟎

DECISION RULE:
𝑹𝑬𝑱𝑬𝑪𝑻 𝑯𝒐 𝒊𝒇
𝒕𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 > 𝒕𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳

𝑨𝑪𝑪𝑬𝑷𝑻 𝑯𝒐 𝒊𝒇
𝒕𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 < 𝒕𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳
 Compute for the test statistic
(𝒕𝒄𝒐𝒎𝒑𝒖𝒕𝒆𝒅)

𝑥= − 𝑥> 5.7 − 5.5 0.2 0.2 0.2

𝑡= = = = =
𝑠=> 𝑠 >
0.6> 0.5> 0.036 + 0.016667 0.052667 0.229493
+ > +
𝑛= 𝑛> 10 15
= 0.87
MAKE DECISION
𝑡-./01234 = 0.87 𝑡-5626-78 = 1.714
𝑺𝒊𝒏𝒄𝒆 𝒕𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 < 𝒛𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 , then ACCEPT Ho.

CONCLUSION
There is not enough evidence to support the claim that taking 40 mg Gingko 3 times
a day for 6 weeks is effective in increasing the mean performance on the Wechsler
Memory Scale at 0.05 level of significance
 DO THIS!
A health psychologist is looking into the effects of different kinds of exercise
on stress. She divides volunteers into two exercise groups: High-Intensity
Interval Training (HIIT) group and the Yoga group. The volunteers participate in
a set number of hours of the exercise for a month, after which the psychologist
measures their stress levels on a scale of 1 – 40 where lower scores mean
lower stress and higher scores mean higher stress. The scores for each group
are shown in the table below

HIIT YOGA
n 8 8
Mean 17.6 20.6
Std Deviation 5.9 4.4
Is there a significant difference between the stress level of
the HIIT and Yoga group? Test at 0.05 level of significance
assuming variances are equal.
Paired t-test
 The paired sample t-test
sometimes called the dependent
sample t-test, is a statistical
procedure used to determine
whether the mean difference
between two sets of observation
is zero. In a paired sample t-test,
each subject or entity is
measured twice, resulting in pair
observations.
 FORMULA
𝑑̅
𝑡=
𝑠32 / 𝑛
Where:
𝜇;: - is the population mean of the difference of the paired data

𝑑̅ - is the mean value of the difference of the paired sample data

𝑠;: - is the standard deviation of the difference of the paired sample data

n - is the size of the sample

 EXAMPLE SUBJECT PRE-TEST POST-TEST

A teacher developed a module that 1 18 22

she claimed would be able to help 2 21 25
increase the problem-solving skills of 3 16 17
her students in Geometry. The 4 22 24
students’ problem-solving skills were 5 19 16
measured and collected from a 6 24 29
selected sample both before and after 7 17 20
using the module. Is there enough 8 21 23
evidence to support the claim of the 9 23 19
teacher that her module can increase 10 18 20
the problem-solving skills of her 11 14 15
students? Test at 0.05 level of 12 16 15
significance. 13 16 18
14 19 26
STATE Ho and Ha: 15 18 18
16 20 24
Ho: 𝝁𝒅 = 𝟎 17 12 18
Ha: 𝝁𝒅 > 𝟎 18 22 25
19 15 19
(One-Tailed Test) 20 17 16
 Identify Significance Level (∞)
and Degrees of Freedom (df)

∝= 𝟎. 𝟎𝟓
d𝒇 = 𝒏 − 𝟏
d𝒇 = 𝟐𝟎 − 𝟏
d𝒇 = 𝟏𝟗
Determine critical value for
(𝒕𝒄𝒓𝒊𝒕𝒊𝒄𝒂𝒍 )
(𝐓𝐡𝐞 𝒕 − 𝒅𝒊𝒔𝒕𝒓𝒊𝒃𝒖𝒕𝒊𝒐𝒏 𝒘𝒊𝒍𝒍 𝒃𝒆 𝒖𝒔𝒆𝒅)

𝑺𝒊𝒏𝒄𝒆:
𝒐𝒏𝒆 − 𝒕𝒂𝒊𝒍𝒆𝒅
∝= 𝟎. 𝟎𝟓
𝒅𝒇 = 𝟏𝟗
𝒕𝒄𝒓𝒊𝒕𝒊𝒄𝒂𝒍 = 𝟏. 𝟕𝟐𝟗
DECISION RULE:
𝑹𝑬𝑱𝑬𝑪𝑻 𝑯𝒐 𝒊𝒇 𝒕𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 > 𝒕𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳
𝑨𝑪𝑪𝑬𝑷𝑻 𝑯𝒐 𝒊𝒇 𝒕𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 < 𝒕𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳
41
COMPUTATION (𝒕𝒄𝒐𝒎𝒑𝒖𝒕𝒆𝒅) 𝑑̅ = = 2.05
20
PRE-TEST POST-TEST $
SUBJECT 𝒅 (𝒅 − 𝒅) (𝒅 − 𝒅 ̅)𝟐
1 18 22 4 1.95 3.8025 ∑(𝑑 − 𝑑)̅ " 152.95
𝑆$# = = = 8.05 = 2.84
2 21 25 4 1.95 3.8025 𝑛−1 19
3 16 17 1 -1.05 1.1025
4 22 24 2 -.05 0.0025 𝑑̅ 2.05
5 19 16 -3 -5.05 25.5025 𝑡!"#$%&'( = = = 3.23
𝑆() / 𝑛 2.84/ 20
6 24 29 5 2.95 8.7025
7 17 20 3 0.95 0.9025
8 21 23 2 -.05 .0025
VI. MAKE DECISION
9 23 19 -4 -6.05 36.6025
10 18 20 2 -.05 0.0025 𝑺𝒊𝒏𝒄𝒆 𝒕𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 > 𝒛𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 ,
11 14 15 1 1.05 1.1025 then REJECT Ho.
12 16 15 -1 -3.05 9.3025
13 16 18 2 -0.05 0.0025
Therefore, there is enough evidence
14 19 26 7 4.95 24.5025
15 18 18 0 -2.05 4.2025
to support the claim of the teacher
16 20 24 4 1.95 3.8025 that her module greatly helped in
17 12 18 6 3.95 15.6025 increasing the problem-solving skills
18 22 25 3 0.95 0.9025
of the participants in mathematics.
19 15 19 4 1.95 3.8025
20 17 16 -1 -3.05 9.3025

41 152.95
One-Way Analysis
of Variance
(𝑨𝑵𝑶𝑽𝑨)
 ANALYSIS OF VARIANCE (ANOVA)

is a statistical method used for testing the

null hypothesis that the means of several
populations are equal. The comparison in
means of 3 or more samples/populations
which follow normal distributions can be
taken simultaneously in just one application
for this test. This test, therefore is a
generalization of the z and t tests of two
normal population means.
 ASSUMPTIONS
1. The sample must be randomly selected
from normal populations.
2. The populations should have equal
variances.
3. The distance from one value to its group’s
mean should be independent of the
distances of other values to that mean.
4. The different samples are from
populations that are categorized in only
one way.
 SAMPLE PROBLEM 1
1. Researchers wanted to test a new State Ho and Ha:
anti-anxiety medication. They split (Since three groups are being compared)
participants into three conditions
Ho: 𝜇7 = 𝜇8 = 𝜇9
(0mg, 50mg and 100mg), then ask
them to rate their anxiety level on a Ha: 𝜇7 ≠ 𝜇8 ≠ 𝜇9
scale of 1-10. Are there any
differences between the three Identify the Level of Significance:
conditions at 0.05 level of
significance? α=0.05
0mg 50mg 100mg Find the critical region (𝑭𝒄𝒓𝒊𝒕𝒊𝒄𝒂𝒍 )
9 7 4 To find the critical region solve for degrees of
freedom (between and within)
8 6 3 𝑑𝑓(+'&,''-) = 𝑛!"/%#- − 1 𝑑𝑓(,0&10-) = 𝑛&"&2/ − 𝑛!"/%#-
7 6 2 𝑑𝑓(+'&,''-) = 3 − 1 𝑑𝑓(,0&10-) = 21 − 3
8 7 3 𝑑𝑓(+'&,''-) = 2 𝑑𝑓(,0&10-) = 18

8 8 4 𝑑𝑓(@A=BAAC) 2
𝐹(:;<=<:>?) = = = 𝟑. 𝟓𝟔
9 7 3 𝑑𝑓(B<=D<C) 18
8 6 2
𝐶𝑙𝑖𝑐𝑘 ℎ𝑒𝑟𝑒 𝑡𝑜 𝑓𝑖𝑛𝑑 𝑜𝑢𝑡 ℎ𝑜𝑤 𝑡𝑜 𝑔𝑒𝑡 3.56
DECISION RULE
𝑹𝑬𝑱𝑬𝑪𝑻 𝑯𝒐 𝒊𝒇 𝑭𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 > 𝑭𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳
𝑨𝑪𝑪𝑬𝑷𝑻 𝑯𝒐 𝒊𝒇 𝑭𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 < 𝑭𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳
COMPUTATION

𝟐 𝟐 100mg 𝟐
0mg 𝒙𝟏 𝒙𝟏 50mg 𝒙𝟐 𝒙𝟐 𝒙𝟑
𝒙𝟑

9 𝟖𝟏 7 𝟒𝟗 4 𝟏𝟔

8 𝟔𝟒 6 𝟑𝟔 3 𝟗

7 𝟒𝟗 6 𝟑𝟔 2 𝟒

8 𝟔𝟒 7 𝟒𝟗 3 𝟗
8 𝟔𝟒 8 𝟔𝟒 4 𝟏𝟔

9 𝟖𝟏 7 𝟒𝟗 3 𝟗

8 𝟔𝟒 6 𝟑𝟔 2 𝟒

L 𝒙𝟏 = 𝟓𝟕 O 𝒙𝟏 𝟐 = 𝟒𝟔𝟕 L 𝒙𝟐 = 𝟒𝟕 O 𝒙𝟐 𝟐 = 𝟑𝟏𝟗 L 𝒙𝟑 = 𝟐𝟏 O 𝒙𝟑 𝟐 = 𝟔𝟕
STEP 1 Solve for TOTAL SUM OF SQUARES (SStotal)

∑ 𝑥 +
𝑆𝑆737BC = M 𝑥+ −
𝑛DEDFG

M 𝑥 = M 𝑥* + M 𝑥+ + M 𝑥H M 𝑥 + = M 𝑥* + + M 𝑥+ + + M 𝑥H +

M 𝑥 = 57 + 47 + 21 = 𝟏𝟐𝟓 M 𝑥 + = 467 + 319 + 67 = 𝟖𝟓𝟑

+ (125) +
∑ 𝑥
𝑆𝑆737BC = M 𝑥+ − = 853 − = 853 − 744.0476 = 108.9524
𝑛DEDFG 21

STEP 2 Solve for SUM OF SQUARES WITHIN (SSWITHIN)

∑ 𝑥* + + ∑ 𝑥+ + + ∑ 𝑥H +
𝑆𝑆IJ7KJL = M 𝑥+ −
𝑛MENO

(57)+ +(47)+ +(21)+ 5899

𝑆𝑆IJ7KJL = 853 − = 853 − = 853 − 842.7143 = 10.2857
7 7
 Solve for
STEP 3 SUM OF SQUARES BETWEEN (SSBETWEEN)

𝑆𝑆P87I88L = 𝑆𝑆737BC − 𝑆𝑆IJ7KJL

𝑆𝑆P87I88L = 108.9524 − 10.2857

𝑆𝑆P87I88L = 108.9524 − 10.2857
𝑆𝑆P87I88L = 98.6667

Solve for
STEP 4 MEAN OF SQUARES (MS)

𝑆𝑆P87I88L 𝑆𝑆IJ7KJL
𝑀𝑆P87I88L = 𝑀𝑆IJ7KJL =
𝑑𝑓P87I88L 𝑑𝑓IJ7KJL

98.6667 10.2857
𝑀𝑆P87I88L = 𝑀𝑆IJ7KJL =
2 18

𝑀𝑆P87I88L = 49.33335 𝑀𝑆IJ7KJL = 0.57143

 CONSTRUCT ANOVA SUMMARY TABLE AND SOLVE FOR F-
STEP 5 STATISTIC (𝑭𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 )

SS df MS 𝑭𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫
𝑴𝑺𝑩𝑬𝑻𝑾𝑬𝑬𝑵
Between 98.6667 2 49.33335 𝑭𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 =
𝑴𝑺𝑾𝑰𝑻𝑯𝑰𝑵

𝟒𝟗. 𝟑𝟑𝟑𝟑𝟓
Within 10.2857 18 0.57143 𝑭𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 =
𝟎. 𝟓𝟕𝟏𝟒𝟑

𝑭𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = 𝟖𝟔. 𝟑𝟑
Total 108.9524 20

MAKE A DECISION
𝑭𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = 𝟖𝟔. 𝟑𝟑 𝑭𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 = 𝟑. 𝟓𝟔

Since, 𝑭𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 > 𝑭𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 Then, REJECT Ho

Therefore, there is a significant difference
CONCLUSION between the three conditions at 0.05 level of
significance
 SAMPLE PROBLEM 1
State Ho and Ha:
2. Suppose the National Transportation (Since three groups are being compared)
Safety Board (NTSB) wants to examine
the safety of compact cars, midsize Ho: 𝜇7 = 𝜇8 = 𝜇9
cars, and full-size cars. It collects a
sample for each of the treatments (car Ha: 𝜇7 ≠ 𝜇8 ≠ 𝜇9
types). Using the hypothetical data
provided below, test whether the mean Identify the Level of Significance:
pressure applied to the driver’s head α=0.01
during a crash test is equal for each
types of car. Use alpha = 0.01 Find the critical region (𝑭𝒄𝒓𝒊𝒕𝒊𝒄𝒂𝒍 )
To find the critical region solve for degrees of
COMPACT MIDSIZE FULL-SIZE freedom (between and within)
CARS CARS CARS 𝑑𝑓(+'&,''-) = 𝑛!"/%#- − 1 𝑑𝑓(,0&10-) = 𝑛&"&2/ − 𝑛!"/%#-
643 469 484 𝑑𝑓(+'&,''-) = 3 − 1 𝑑𝑓(,0&10-) = 15 − 3
655 427 456 𝑑𝑓(+'&,''-) = 2 𝑑𝑓(,0&10-) = 12
702 525 402
𝑑𝑓(@A=BAAC) 2
685 512 445 𝐹(:;<=<:>?) = = = 𝟔. 𝟗𝟑
625 451 552 𝑑𝑓(B<=D<C) 12

𝐶𝑙𝑖𝑐𝑘 ℎ𝑒𝑟𝑒 𝑡𝑜 𝑓𝑖𝑛𝑑 𝑜𝑢𝑡 ℎ𝑜𝑤 𝑡𝑜 𝑔𝑒𝑡 6.93

 DECISION RULE
𝑹𝑬𝑱𝑬𝑪𝑻 𝑯𝒐 𝒊𝒇 𝑭𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 > 𝑭𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳
𝑨𝑪𝑪𝑬𝑷𝑻 𝑯𝒐 𝒊𝒇 𝑭𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 < 𝑭𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳
COMPUTATION
COMPACT MIDSIZE FULL-SIZE
CARS 𝟐 CARS 𝟐 CARS 𝟐
𝒙𝟏 𝒙𝟐 𝒙𝟑
(𝒙𝟏 ) (𝒙𝟐 ) (𝒙𝟑 )
643 𝟒𝟏𝟑, 𝟒𝟒𝟗 469 𝟐𝟏𝟗, 𝟗𝟔𝟏 484 𝟐𝟑𝟒, 𝟐𝟓𝟔

655 𝟒𝟐𝟗, 𝟎𝟐𝟓 427 𝟏𝟖𝟐, 𝟑𝟐𝟗 456 𝟐𝟎𝟕, 𝟗𝟑𝟔

702 𝟒𝟗𝟐, 𝟖𝟎𝟒 525 𝟐𝟕𝟓, 𝟔𝟐𝟓 402 𝟏𝟔𝟏, 𝟔𝟎𝟒

685 𝟒𝟔𝟗, 𝟐𝟐𝟓 512 𝟐𝟔𝟐, 𝟏𝟒𝟒 445 𝟏𝟗𝟖, 𝟎𝟐𝟓

625 𝟑𝟗𝟎, 𝟔𝟐𝟓 451 𝟐𝟎𝟑, 𝟒𝟎𝟏 552 𝟑𝟎𝟒, 𝟕𝟎𝟒

L 𝒙𝟏 = 𝟑𝟑𝟏𝟎 L 𝒙𝟏 𝟐 = 𝟐, 𝟏𝟗𝟓, 𝟏𝟐𝟖 L 𝒙𝟐 = 𝟐𝟑𝟖𝟒 L 𝒙𝟐 𝟐 = 𝟏, 𝟏𝟒𝟑, 𝟒𝟔𝟎 L 𝒙𝟑 = 𝟐𝟑𝟑𝟗 L 𝒙𝟑 𝟐 = 𝟏, 𝟏𝟎𝟔, 𝟓𝟐𝟓

STEP 1 Solve for TOTAL SUM OF SQUARES (SStotal)

∑ 𝑥 +
𝑆𝑆737BC = M 𝑥+ − M 𝑥 + = M 𝑥* + + M 𝑥+ + + M 𝑥H +
𝑛DEDFG

M 𝑥 = M 𝑥* + M 𝑥+ + M 𝑥H M 𝑥 + = 2,195,128 + 1,143,460

+1,106,525 = 𝟒, 𝟒𝟒𝟓, 𝟏𝟏𝟑

M 𝑥 = 3310 + 2384 + 2339 = 𝟖, 𝟎𝟑𝟑

∑𝑥 + (8,033)+
𝑆𝑆737BC = M 𝑥+ − = 4,445,113 − = 4,445,113 − 4,301,939.2667
𝑛DEDFG 15
𝑆𝑆737BC = 143,173.7333

STEP 2 Solve for SUM OF SQUARES WITHIN (SSWITHIN)

∑ 𝑥* + + ∑ 𝑥+ + + ∑ 𝑥H +
𝑆𝑆IJ7KJL = M 𝑥+ −
𝑛MENO

(3310)+ +(2384)+ +(2339)+ 22,110,477

𝑆𝑆IJ7KJL = 4,445,113 − = 4,445,113 −
5 5
𝑆𝑆IJ7KJL = 4,445,113 − 4,422,095.4 = 𝟐𝟑, 𝟎𝟏𝟕. 𝟔
 Solve for
STEP 3 SUM OF SQUARES BETWEEN (SSBETWEEN)

𝑆𝑆P87I88L = 𝑆𝑆737BC − 𝑆𝑆IJ7KJL

𝑆𝑆P87I88L = 143,173.7333 − 23,017.6

𝑺𝑺𝑩𝑬𝑻𝑾𝑬𝑬𝑵 = 𝟏𝟐𝟎, 𝟏𝟓𝟔. 𝟏𝟑𝟑𝟑

Solve for
STEP 4 MEAN OF SQUARES (MS)
𝑆𝑆P87I88L 𝑆𝑆IJ7KJL
𝑀𝑆P87I88L = 𝑀𝑆IJ7KJL =
𝑑𝑓P87I88L 𝑑𝑓IJ7KJL

120,156.1333 23,017.6
𝑀𝑆P87I88L = 𝑀𝑆IJ7KJL =
2 12

𝑴𝑺𝑩𝑬𝑻𝑾𝑬𝑬𝑵 = 𝟔𝟎, 𝟎𝟕𝟖. 𝟎𝟔𝟔𝟔𝟓 𝑴𝑺𝑾𝑰𝑻𝑯𝑰𝑵= 𝟏, 𝟗𝟏𝟖. 𝟏𝟑𝟑𝟑𝟑

 CONSTRUCT ANOVA SUMMARY TABLE AND SOLVE FOR F-
STEP 5 STATISTIC (𝑭𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 )

SS df MS 𝑭𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫
𝑴𝑺𝑩𝑬𝑻𝑾𝑬𝑬𝑵
Between 𝟏𝟐𝟎, 𝟏𝟓𝟔. 𝟏𝟑𝟑𝟑 2 𝟔𝟎, 𝟎𝟕𝟖. 𝟎𝟔𝟔𝟔𝟓 𝑭𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 =
𝑴𝑺𝑾𝑰𝑻𝑯𝑰𝑵

𝟔𝟎, 𝟎𝟕𝟖. 𝟎𝟔𝟔𝟔𝟓

Within 23,017.6 12 𝟏, 𝟗𝟏𝟖. 𝟏𝟑𝟑𝟑𝟑 𝑭𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 =
𝟏, 𝟗𝟏𝟖. 𝟏𝟑𝟑𝟑𝟑

𝑭𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = 𝟑𝟏. 𝟑𝟐
Total 143,173.73333 14

MAKE A DECISION
𝑭𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = 𝟑𝟏. 𝟑𝟐 𝑭𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 = 𝟔. 𝟗𝟑

Since, 𝑭𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 > 𝑭𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 Then, REJECT Ho

Therefore, the mean pressure applied to the
CONCLUSION driver’s head during a crash test is not equal for
each types of car at 0.01 level of significance
𝑀𝐴𝐾𝐸 𝑆𝑈𝑅𝐸 𝑇𝐻𝐴𝑇 𝑌𝑂𝑈 𝑈𝑆𝐸 𝑇𝐻𝐸 𝑅𝐼𝐺𝐻𝑇 𝑇𝐴𝐵𝐿𝐸. 𝐼𝑁 𝑇𝐻𝐼𝑆 𝐶𝐴𝑆𝐸 𝐹𝑂𝑅 𝛼 = 0.05
2 𝑙𝑜𝑜𝑘 𝑓𝑜𝑟 2 𝑖𝑛 𝑡ℎ𝑒 𝑢𝑝𝑝𝑒𝑟 𝑝𝑎𝑟𝑡
𝛼 = 0.05 𝑑𝑓 =
18 𝑙𝑜𝑜𝑘 𝑓𝑜𝑟 18 𝑖𝑛 𝑡ℎ𝑒 𝑙𝑒𝑓𝑡𝑚𝑜𝑠𝑡 𝑝𝑎𝑟𝑡

𝑭𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 = 𝟑. 𝟓𝟔 𝑪𝒍𝒊𝒄𝒌 𝒕𝒉𝒊𝒔 𝒕𝒐 𝒓𝒆𝒕𝒖𝒓𝒏

𝑀𝐴𝐾𝐸 𝑆𝑈𝑅𝐸 𝑇𝐻𝐴𝑇 𝑌𝑂𝑈 𝑈𝑆𝐸 𝑇𝐻𝐸 𝑅𝐼𝐺𝐻𝑇 𝑇𝐴𝐵𝐿𝐸. 𝐼𝑁 𝑇𝐻𝐼𝑆 𝐶𝐴𝑆𝐸 𝐹𝑂𝑅 𝛼 = 0.01

2 𝑙𝑜𝑜𝑘 𝑓𝑜𝑟 2 𝑖𝑛 𝑡ℎ𝑒 𝑢𝑝𝑝𝑒𝑟 𝑝𝑎𝑟𝑡

𝛼 = 0.01 𝑑𝑓 =
12 𝑙𝑜𝑜𝑘 𝑓𝑜𝑟 12 𝑖𝑛 𝑡ℎ𝑒 𝑙𝑒𝑓𝑡𝑚𝑜𝑠𝑡 𝑝𝑎𝑟𝑡

𝑭𝑪𝑹𝑰𝑻𝑰𝑪𝑨𝑳 = 𝟔. 𝟗𝟑 𝑪𝒍𝒊𝒄𝒌 𝒕𝒉𝒊𝒔 𝒕𝒐 𝒓𝒆𝒕𝒖𝒓𝒏