HYPOTHESIS TESTING

Statistical Hypothesis is an assumption or statement concerning a

characteristic of a population.

Two Types of Statistical Hypothesis

1. Null Hypothesis (Ho)

 it specifies the statement to be tested.
 assumed to be true throughout the statistical analysis.
 it is a hypothesis of no difference.

2. Alternative Hypothesis (H1)

 it specifies the values which are to be accepted if Ho is rejected.

Two Classification of Alternative Hypothesis

1. Non-directional Hypothesis – makes use of “not equal to” (  ) sign.
2. Directional Hypothesis – makes use of “less than” (<) or “greater
than” (>).

Example 1

Formulate the null (Ho) and alternative (H1) hypotheses for the following
problems and identify if the hypothesis is directional or non-directional.

1. A doctor wants to know if the average recovery time of a patient taking a

particular medication is one month.

Ho :   1 (The average recovery time of a patient taking a particular

medication is one month.)
H1:   1 (The average recovery time of a patient taking a
particular medication is not one month.)

Classification: Non-Directional

2. A researcher wants to determine if there is a significant difference of the

performances of the faculty members between the two colleges.

Ho : 1   2 (There is no significant difference of the performances of

the faculty members between the two colleges.)
H1: 1   2 (There is a significant difference of the performances
of the faculty members between the two colleges.)

Classification: Non-Directional
3. A drug store wants to know if the average number of sold paracetamol
tablets is more than 100 per day.

Ho :   100 (The average number of sold paracetamol tablets is

100 per day.)
H1:   100 (The average number of sold paracetamol tablets is
greater than 100 per day.)

Classification: Directional

4. An internet provider company X claims that their internet connection is

faster than the other company Y.

Ho: 1   2 (There is no significant difference of the internet

connection between the two companies X and Y.)
H1: 1   2 (Company X provides faster internet connection than
company Y.)

Classification: Directional

Try this!

Direction: Formulate the null (Ho) and alternative (H1) hypotheses for the
following problems and identify if the hypothesis is directional or
non-directional.

1. A clothing store owner believes that the mean income of its customers is
Php 20,000 per month.

2. An automobile manufacturer states that the mean kilometer per gallon

(kpg) for one of its new full-sized models is less than 60.

3. Recorded speech can be compressed and played back at a faster rate. A

paper gave the results of a study design to test comprehension of
time-compressed speech. A researcher wants to know if the mean
comprehension scores for the students hearing a time-compressed lecture
is significantly lower than the mean score for students who hear a lecture
at a normal speed.
4. The manufacturer of a gasoline additive claims that the use of this additive
increases gasoline mileage than without additives.

Possibilities in the Decision Procedure

• The null hypothesis is accepted such that it is true.

• The null hypothesis is accepted when in fact, it is false.
• The null hypothesis is rejected when in fact, it is true.
• The null hypothesis is rejected such that it is false.

We arrived at a “PERFECT DECISION” if we accept either possibility 1 or 4.

We commit an “ERROR” if we accept either possibility 2 or 3.

Two Types of Error

1. Type I Error – The null hypothesis is rejected when in fact it is true.

2. Type II Error – The null hypothesis is accepted when in fact it is false.

Table 10.1: Consequences of Decisions in Testing Hypothesis

Decision/Fact H0 is True H0 is false

ACCEPT H0 Correct Decision Type II Error

REJECT H0 Type I Error Correct Decision

Moreover, the probability of occurrence of these events is denoted by:

P(Type 1 Error) = 
P(Type 2 Error) = 

In testing hypothesis, we call  as the level of significance, which is the

maximum probability with which we are willing to take a risk of committing
a Type I Error.

Example 2

If  㺘☮㺘ٝ , then there are about 5 chances in 100 that we reject the null
hypothesis when it should be accepted. We are also 95% confident that we
made the right decision.

Two Types of Tests

 1. one-tailed test (one-sided test)
 A test of any statistical hypothesis where H1 is directional.
 Characterized by a region of rejection which lies entirely in one end
of the distribution.

2. two-tailed test (two-sided test)

 A test where H1 is non-directional.
 Involves a critical region which is split into two equal areas placed
in each tail of the distribution.

10.1 Test for Population Means

 Test Statistic - the test used to determine whether Ho should be

rejected or not.
 Critical Value - identifies the value of the test statistic that would
lead to the rejection of Ho at a chosen level of significance.
 Critical Region (rejection region) - is a part of the set of all possible
values of a test statistic for which Ho is rejected.

Steps in Hypothesis Testing

1. Formulate the null hypothesis and the alternative hypothesis.

2. Specify the level of significance, .
3. Choose the appropriate test statistic.
4. Establish the critical region/regions.
5. Compute the actual value of the test statistic from the sample.
6. Make the statistical decision, that is, reject the null hypothesis when
the computed value of the test statistic is within the critical region;
otherwise accept Ho.
7. Draw the conclusion.

Table 10.2: Testing One Population Mean: H0 :   0

H0 Test Statistics H1 Critical Regions

if  is known or 㺘,   0 z  z 

z
x   
0 n   0 z  z
   0 z   z or z  z 
(i)   0 2 2

z
x   0 n
s
 if  is unknown and t 㺘,   0 t  t ,v

(ii)    0 t=
x    0 n
,
  0 t  t ,v
s   0 t  t
,v
or t  t
,v
2 2
with degree of freedom, ᮐ

Example 3

A frozen food company wishes to know the mean length of corn received in a
large shipment. A random sample of 20 corns were collected and measured
and found to have a mean length of 8.0 inches. It has been known that the
standard deviation of all corn shipments is 1.5 inches. Is there a reason to
believe that the mean length of the corn is not 9.0 inches at 㺘☮㺘ٝ?

Solution: Given: h㺘 ☮ٝ
t☮㺘 㺘☮㺘ٝ
1. H 0 :   9.0
H1 :   9.0 (two-tailed)

2. 㺘☮㺘ٝ

3.
z
x    0 n


4. Looking at the table (critical values for Z) on the last page of this module,
z   z z  z
2 or 2

z   z0.025 or z  z0.025
z  1.96 or z  1.96 (critical region)

t☮㺘ᮐ❨☮㺘ٝ h㺘
5. ☮ٝ
ᮐ h☮❨t

6. z  1.96 or z  1.96
ᮐ h☮❨t tᮐ ☮❨
 since the calculated z-value is within the critical region, we
Reject H0

7. The mean length of the corn is not 9.0 inches.

Example 4
The average weight of 28 randomly selected sacks of rice is 48.54 kilograms
with a standard deviation of 20 kilograms. Test the hypothesis at 0.01 level of
significance that the true mean weight is less than 50 kilos.

Solution: Given: ht h㺘
ot☮ٝo 㺘☮㺘

1. H 0 :   50
H 1 :   50 (one-tailed)

2.   0.01

3. t =
x    0 n
s

4. tᮐ
Degree of freedom ᮐ ht ᮐ h௲
tᮐ 㺘☮㺘
Looking at the t-distribution table on the last page of this module (one-tailed,
  0.01 , degree of freedom =27), we have
tᮐ )☮Հ
critical regionٝ

ot☮ٝoᮐٝ㺘ٝ ht
5. h㺘
ᮐ 㺘☮ t

6. tᮐ h☮o௲
ᮐ 㺘☮ t ᮐ h☮o௲
 since the calculated value is NOT WITHIN the critical value , we
Accept H 0

7. The true mean weight is 50 kilos.

Try this!

1. The treasurer of a certain university claims that the mean monthly salary of
their college professor is Php21,750.00 with a standard deviation of
Php6,000.00. A researcher takes a random sample of 75 college professor and
were found to have a mean monthly salary of Php19,375.00. Do the 75 college
professors have a lower salary than the rest? Test the claim at 0.05 level of
significance.
2. A nationwide survey found out that the average time that college students
spent on their personal computer is 7.5 hours per week. A random sample of
28 college students showed that they spent 7.1 hours per week using their
computers with a standard deviation of 0.7 hours. Test whether the average
number of hours spent by the 28 college students is significantly lower than
the national average of 7.5 hours. Use a 0.05 level of significance.
Critical Values for z:

=0.01 =0.05
ᮐ h☮ ☮ oٝ
‫ ݓ‬ᮐ h☮ٝ௲ٝ ☮❨

t-distribution table:

Degree of freedom