HYPOTHESIS

TESTING

FOR LECTURE PURPOSES ONLY MANNY MACAM

What is hypothesis?
• Hypothesis is a predictive statement, capable
of being tested by scientific methods, that
relates an independent variables to some
dependent variable.
• A hypothesis states what we are looking for
and it is a proportion which can be put to a
test to determine its validity
What is hypothesis?
Example
Students who receive counseling will show a
greater increase in creativity than students
not receiving counseling.
Characteristics of Hypothesis
 Clear and precise.
 Capable of being tested.
 Stated relationship between variables.
 Limited in scope and must be specific
 Stated as far as possible in most simple
terms so that the same is easily understand
by all concerned (But one must remember
that simplicity of hypothesis has nothing to
do with its significance).
Characteristics of Hypothesis
 Consistent with most known facts.
 Responsive to testing with in a reasonable
time. One can’t spend a life time collecting
data to test it.
 Explain what it claims to explain; it should
have empirical reference
Null Hypothesis
 It is an assertion that we hold as true unless
we have sufficient statistical evidence to
conclude otherwise.
 Null Hypothesis is denoted by 𝐻0
 If a population mean is equal to hypothesized
mean then Null Hypothesis can be written as
𝐻0 : 𝜇 = 𝜇 0
Alternative Hypothesis
 The Alternative hypothesis is negation of null
hypothesis and is denoted by 𝐻𝑎
 If Null Hypothesis is given as
𝐻0 : 𝜇 = 𝜇 0

Then Alternative Hypothesis can be written

as 𝐻𝑎 : 𝜇  𝜇0
𝐻𝑎 : 𝜇 > 𝜇 0
𝐻𝑎 : 𝜇 < 𝜇 0
Level of significance and confidence
 Significance means the percentage risk to
reject a null hypothesis when it is true and it
is denoted by 𝛼
 Generally taken as 1%, 5%, 10%
 (1 − 𝛼) is the confidence interval in which the
null hypothesis will exist when it is true.
Type I and Type II Error

DECISION
DECISION
SITUATION
SITUATION
ACCEPT
ACCEPT NULL
NULL REJECT
REJECT NULL
NULL

Type
Type II Error
Error
Null is true Correct
Correct
((𝛼 error)
error)

Type
Type IIII Error
Error
Null is false Correct
Correct
(( error)
error)
Two-tailed test @ 5% Significance
level Acceptance and Rejection
regions in case of a Two-tailed
  Suitable when 𝐻00test
: 𝜇 = 𝜇00 and 𝐻𝑎𝑎 : 𝜇
𝜇00
𝑅𝑒𝑗𝑒𝑐𝑡𝑖𝑜𝑛 𝑅𝑒𝑗𝑒𝑐𝑡𝑖𝑜𝑛
𝑟𝑒𝑔𝑖𝑜𝑛 𝑟𝑒𝑔𝑖𝑜𝑛
/𝑠𝑖𝑔𝑛𝑖𝑓𝑖𝑐𝑎𝑛𝑐𝑒 Total /𝑠𝑖𝑔𝑛𝑖𝑓𝑖𝑐𝑎𝑛𝑐𝑒
𝑙𝑒𝑣𝑒𝑙 (𝛼 = 𝐴𝑐𝑐𝑒𝑝𝑡𝑎𝑛𝑐𝑒 𝑙𝑒𝑣𝑒𝑙 (𝛼 =
0.025 𝑜𝑟 2.5%) 𝑟𝑒𝑔𝑖𝑜𝑛 𝑜𝑟 0.025 𝑜𝑟 2.5%)
𝑐𝑜𝑛𝑓𝑖𝑑𝑒𝑛𝑐𝑒 𝑙𝑒𝑣𝑒𝑙
(1 − 𝛼) = 95%
𝐻0 : 𝜇 = 𝜇0
Left-tailed test @ 5% Significance
level Acceptance and Rejection
regions in case of a left-tailed
   Suitable when 𝐻
test
00
: 𝜇 = 𝜇00 and 𝐻𝑎𝑎 :
𝜇 𝜇00
𝑅𝑒𝑗𝑒𝑐𝑡𝑖𝑜𝑛
𝑟𝑒𝑔𝑖𝑜𝑛
/𝑠𝑖𝑔𝑛𝑖𝑓𝑖𝑐𝑎𝑛𝑐𝑒 Total
𝑙𝑒𝑣𝑒𝑙 (𝛼 = 0.05 𝑜𝑟 𝐴𝑐𝑐𝑒𝑝𝑡𝑎𝑛𝑐𝑒
5%) 𝑟𝑒𝑔𝑖𝑜𝑛 𝑜𝑟
𝑐𝑜𝑛𝑓𝑖𝑑𝑒𝑛𝑐𝑒 𝑙𝑒𝑣𝑒𝑙
(1 − 𝛼) = 95%
𝐻0 : 𝜇 = 𝜇0
Right-tailed test @ 5% Significance
level Acceptance and Rejection
regions in case of a right-tailed
   Suitable when 𝐻
test
00
: 𝜇 = 𝜇00 and 𝐻𝑎𝑎 :
𝜇 𝜇00
𝑅𝑒𝑗𝑒𝑐𝑡𝑖𝑜𝑛
𝑟𝑒𝑔𝑖𝑜𝑛
Total /𝑠𝑖𝑔𝑛𝑖𝑓𝑖𝑐𝑎𝑛𝑐𝑒
𝐴𝑐𝑐𝑒𝑝𝑡𝑎𝑛𝑐𝑒 𝑙𝑒𝑣𝑒𝑙 (𝛼 = 0.05 𝑜𝑟
𝑟𝑒𝑔𝑖𝑜𝑛 𝑜𝑟 5%)
𝑐𝑜𝑛𝑓𝑖𝑑𝑒𝑛𝑐𝑒 𝑙𝑒𝑣𝑒𝑙
(1 − 𝛼) = 95%
𝐻0 : 𝜇 = 𝜇0
Procedure for Hypothesis Testing
1 State the null (𝐻 ) hypothesis
0
0

2 State the alternate (𝐻 ) Hypothesis

𝑎
𝑎

3 Determine
tailed)
if type off Test (Left-tailed, one-tailed or two-

4 Specify the significance level; 1%, 5%, 10% etc.

5 Decide a test statistics; z-test, t- test, F-test.
Procedure for Hypothesis Testing
6 Use appropriate formula
7 Calculate the value of test statistics.
8 Illustrate
regions)
the tabular value (Acceptance and rejection

9 (accept) H
Make decision on whether to reject or not to reject

1 0.
0.

Form conclusion based from the decision.

0
Procedure for Hypothesis Testing

Compare the p-
value with
calculated value

P-value > calculated P-value < calculated

value value
Accept
Reject 𝐻00
𝐻00
 z-test
HYPOTHESIS
TESTING
When to use z-
test? is less than 30
 Sample
n < 30
 Standard deviation () is known

 Sample is greater or equal than 30

n ≥ 30
 Standard deviation () is either known or
unknown
 t-test
HYPOTHESIS
TESTING
When to use t-
test? is less than 30
 Sample
n < 30
 Standard deviation () is unknown
Test Statistics

If  is known If  is
unknown
Example 1:
A principal at a certain school claims that the students
in his school are above average intelligence. A random
sample of thirty students IQ scores have a mean score
of 112.5. Is there sufficient evidence to support the
principal’s claim? The mean population IQ is 100 with a
standard deviation of 15. IQ’s are normally distributed.
Solution:
Step 1: State the Null Hypothesis.
H0: The accepted fact is that the population mean is
100, so: H0: μ=100
Step 2: State the Alternate Hypothesis.
Ha: The claim is that the students have above average
IQ scores, so: Ha: μ > 100.
Step 3: State the type of test: One-tailed
(or simply right-tailed to be more definite)
Step 4: State the alpha level. 0.05
(Note 0.05 shall be used unless otherwise indicated in the
problem)
Solution:
Step 5: Establish Test statistic. z-
test
Step 6: Use appropriate X - m0
formula z=
s
n
Step 7: Compute Test statistic
value z = 4.56
Solution:
Step 8: Illustrate tabular value. First set our critical
value. It is simpler to determine the z-score from our
SND table. For the significance level  5% in a one-
tailed test is 1.645,
Solution:
Step 8: The critical value z may also be obtained
using the
z-table.

45%

5%

1.645
45% (0.4500) is so far midway between P(z=1.64) =
0.4495 and P(z=1.65) = 0.4505 and we can say that
P(z=1.645) = 0.4500.
Solution:
Step 8: The critical
value z may also be
P(z=1.640) = 0.4495
obtained using the z-
table. P(z=1.645) = 0.4500
P(z=1.64) = 0.4495 P(z=1.650) = 0.4505
and
P(z=1.65) = 0.4505
We can say that
Z = 1.64 Z = 1.65
P(z=1.645) =
0.4500.
Solution:
Step 8: Acceptance 4.56 lies to the
Hence, region right of the
critical value
1.645
Rejection
region

1.645
4.56

NOTE: In this case, the computed value zc it is

greater
(4.56 > 1.645). Thus, it lies on the rejection
region
Solution:
Step 9: Establish decision. Since 4.56 > 1.645.
Reject the null hypothesis. That is Reject H0.

Step 10: Make conclusion: The claim that the

students are above average
intelligence is correct
Example 2:
The price of a popular tennis racket at a national chain
 

store is $179. Portia bought five of the same racket at

an online auction site for the following prices:
$155 $179 $175 $175 $161
Assuming that the auction prices of rackets are
normally distributed, determine whether there is
sufficient evidence in the sample, at the 5% level of
significance, to conclude that the average price of the
racket is less than $179 if purchased at an online
auction.

Note: Computed (s=10.39) is and mean (=169)

Example 2:
Establish the given
data: 
n=5
df = n -1 =4
=169 (Average of the 5 data)
0=179
s= 10.39 (Computed from calculator)
Solution:
Step 1: State the Null Hypothesis.
H0: μ= μ0  H0: μ=179

Step 2: State the Alternate Hypothesis.

Ha: μ<μ0  H0: μ<179

Step 3: Left-tailed
Step 4: Level of significance 0.05
Step 5: Test statistic. t-test
[Since n=5 (n<30), so t-test is used.]
Solution:
Step 6: Use appropriate
formula

Step 7: Compute Test statistic t = -2.152

value.
169 - 179
z= =- 2.152
10.39
5
Solution:
Step 8: Illustrate tabular value. The critical value
from the t-tabe gives is −2.1318 (left-tailed, use
negative value).

t = 2.1318
Solution:
Acceptance 4.56 lies to the
region left of the
critical value
Rejection 1.645
region

−2.1318
-2.152

NOTE: In this case, the computed value zc it is less than

the critical value, (-2.152 < −2.1318). Thus, it lies on the
rejection region
Solution:
Step 9: Establish decision. Since (-2.152 <
−2.1318), we reject the null hypothesis. That is
Reject H0.
Step 10: Make conclusion: that the average price of
the racket is less than $179 if purchased at an
online auction.
SAMPLEPROBLEMS
Sample problems
1. A manufacturing process of produces ball bearings
with diameters that have a normal distribution with
known standard deviation of 0.04 cm. Ball bearings
with diameters that are too small or too large are
undesirable. In order to test the claim that the
population mean is 0.50, perform a two–tailed
hypothesis test at the 5% level of significance. Assume
that a random sample of 25 gave a mean diameter of
0.51 cm.
Sample problems
2. The manufacturer of electric bulb has established
that the average life of the bulb that his firm produces
is 800 hours with a standard deviation of 40 hours. The
production of bulbs are closely monitored so that the
bulb lifetime are maintained close to the average.
Occasionally, he takes a random sample bulbs to see if
the average lifetime of bulbs may have significantly
changed. On one particular occasion, he took a sample
of 30 bulbs and found that the average to be 788
hours. At 10% level of significance, does the
manufacturer have reasons to believe that the average
lifetime of bulbs manufactured by his company has
decreased?
Sample problems
3. In the past, the average length of outgoing
telephone calls from a business office has been 143
seconds. A manager wishes to check whether that
average has decreased after the introduction of new
policies. A sample of 100 telephone calls produced a
mean off 133 seconds with a standard deviation of 3
seconds. Perform the relevant test of hypothesis using
1% level of significance to test if the length of average
calls have decreased.
Sample problems
4. The average number of days to complete a recovery
from a particular knee operation is 123.7 days. From
experience, a Physician suspects that the use of topical
pain medication might be lengthening the recovery
time. He randomly selects the records of 1 surgery
patients who used the topical medication obtaining a
mean of 132.29 and a standard deviation of 10.98.
Assuming a normal distribution of recovery time,
perform the relevant test of hypothesis at 5% level of
significance.