CHAPTER

THREE
HYPOTHESIS
1
TESTING
BASIC CONCEPTS OF HYPOTHESIS

 Hypothesis: A statement about the value of a population parameter.

 Hypothesis Testing A procedure based on sample evidence and
probability theory used to determine whether the hypothesis is a
reasonable statement and should not be rejected, or is unreasonable
and should be rejected.
 Null Hypothesis: a tentative assumption made about the value of a
population parameter. It is denoted by H0.

 Alternative hypothesis: A statement that will be accepted if our

sample data provide us with sample evidence that the null
2

hypothesis is false. It is denoted by Ha.

STEPS OF HYPOTHESIS TESTING
 In conducting hypothesis testing there are required procedures.
The steps that can be applied to any hypothesis test are:
1. Establish hypotheses.
2. Determine alpha
3. Determine the appropriate statistical test
4. Establish the critical value of the test statistic (state a decision
rule)
5. Gather data
3
6. Calculate the value of the test statistic.
7. Make a decision (reject or accept the hypothesis).
CONT.…
 For our practical purpose, assuming the necessary data is given, we
will use the following four basic steps throughout the material.
1. State the hypotheses.
2. State a decision rule
3. Compute the value of the test statistic.
4. Make a Decision (Reject or Accept the Ho).
5. Conclusion

4
TYPE I AND TYPE II ERRORS
 A conclusion is correct if a true null hypothesis is accepted, or, if a
false null hypothesis is rejected. An error is made under the
following conditions:
 If a true H0 is rejected - Type I error. ( α)

 If a false H0 is accepted - Type II error

 the level of significance, α, is the probability of committing a Type

I error, and is the proportion of the area of the curve occupied by
the rejection region.
5
CONT.….
State of nature

Null (Ho) true Null (Ho) False

Accept Correct decision Type II error (  )

Action

Reject Correct decision

Type I error
 
6
TESTING A RESEARCH HYPOTHESIS
One -tailed and two- tailed test
 One-tailed Test: A hypothesis test in which rejection of the null
hypothesis occurs for values of the test statistic in one tail of the
sampling distribution. This will happen when Ha contains “>” or
“<” sign.
 Two-tailed Test: A hypothesis test in which rejection of the null
hypothesis occurs for values of the test statistic in either tail of the
sampling distribution. This will exist when Ha contains “” sign.

7
CONT.….
Additional concepts on testing hypothesis
 Test statistic: A value, determined from sample information, used
to accept or reject the null hypothesis.
 Critical Value: A number that is the dividing point between the
region of acceptance and the region of rejection.
 It is a value that is compared with the test statistic to determine
whether or not null hypothesis should be rejected.

 Level of significance: The maximum probability of a type I error.

That is, the risk. We assume of rejecting the null hypothesis when
8
it is actual1y true.
CONT.…
 There are three possible null hypotheses about a population mean,
and their corresponding alternative hypotheses

(i)
H 0 :   h (ii)
H 0 :   h (iii)
H 0 :   h

 note that:
So Ha: h Ha:    h Ha:    h
 The first two ((i) and (ii) lead to one-tailed test.
 The first one (i) is a right-tailed test and the 2 nd (ii) is a left-

tailed test.
 The third one (iii) is a two-tailed test.

9
TEST OF A POPULATION MEAN
 While testing populations mean:
 Use the sample Z test statistic when:
 The population is normal and the sample size
whatever but  is known, or,
 The sample size is greater than 30.
 Use the sample t test statistic when the population is
normal,  is unknown, and n <30.
10
TEST OF A POPULATION MEAN
Example 1:
 Suppose that a particular University dean of students has generally
assumed the average age of a student is no more than 20 years.
However, lately the students have appeared to be somewhat older
than before, and the office believes that the average age now might
be older. Suppose that 50 students are chosen from enrollment
records randomly and the mean found to be 20.76. If the
population standard deviation for the ages of these university
students is 3.6 years, perform a hypothesis test at α = 0.05.
11
CONT.…
Solution:
 Step 1; State the hypothesis:

Ho: µ 20 years
Ha: µ > 20 years
 Step 2: State the decision rule:
Decision rule will be: Reject Ho if sample Z > 1.645.
b/c 0.5- 0.05 = 0.45 so, z = 1.645 from the table of z-
distribution where z = 45% then find z – value from the normal
z – distribution.
12
CONT.…
 Step 3: computing the test state; Compute the value of the test
statistic (in this case the Z-value):

 Step 4: make decision ; Accept or reject H0:

 Computed (sample) Z > 1.645
1.49 > 1.645, False
13
CONT.…
 Step 5: conclusion
 Thus, the Ho is not rejected. So, Accept the Ho. Although an
average of 20.76 year of age was attained from a random sample
of 50 students, there is no enough evidence to statistically reject a
hypothesized value of 20. The sample mean value of 20.76 years
may be viewed as a mere chance variation of 50 students from a
population with a mean of 20 years of age.

14
CONT.…
Example 2:
 The president of a chain store selling diary products asserts that the
mean content of the store's 32ounce fluid (1 quarter) milk
containers is at least 32 ounce. Performa a hypothesis test at the
1% level of significance if the mean content of a random sample of
60 containers is 31.98 ounces and the sample standard deviation is
0.1 ounce.

15
CONT.…
 Step 1: State the hypotheses:
Ho: µ ≥ 32
Ha: µ < 32
 Step 2: State the decision rule:
 The test will be left-tailed because Ha states is less than32. The
sample size is 60, is greater than 30, so the sample statistic is
the sample Z. thus, Z0.01 = -2.33 (0.5 – 0.01 = 0.49 find 0.49 in
the body of the table and read value of Z)
 Reject H0 if sample Z< -2.33
16
CONT.…
 Step 3 : Compute the sample statistic (Z – value)

 Step 4:
 Accept or reject H0:
 Sample (computed) Z < - 2.33 17

 -1.55 < -2.33 False.

 Step 5: conclusion
 Accept the Ho. Thus, we accept the president's assertion that the
mean content of the containers is at least 32 ounces.

18
CONT.…
Example 4:
 It is known that the distribution of efficiency ratings for production
employees at Muger Cement factory is normally distributed with a
population mean of 200 and a population standard deviation of 16.
The research department is challenging this mean stating it is
different from 200. Using the 0.01 level of significance, test the
hypothesis that the population mean is 200, if the efficiency rating
of 100 employees were analyzed and sample mean is found to be
203.5. 19
Solution
 Step 1: State the hypotheses:
 Ho: µ = 200
 Ha: µ200.
 Step 2: State decision rule:

 Since the alternative hypothesis (Ha) has a symbol, it is a two-tailed

test. In addition, as it is a two-tailed test, half of a, (0.01) that is
0.005, is in each tail. The acceptance region, which is located
between the two tails, is therefore 0.99. Hence, 0.5 - 0.005 = 0.4950
is the area between critical value. Reading this from the body of the
table, the nearest to 0.4950 is 0.4951 in the table; the corresponding
Z value will be 2.58. Thus, the decision rule is:
20
 Reject the Ho if computed (sample) Z value is > 2.58 or < -2.58.
 Step 3: Compute the sample statistic (Z – value)

= 2.19

 Step 4 Accept or Reject Ho:

 /Sample Z/ > 2.58 2.19 > 2.58 False

 Accept the Ho. The population mean is not different from 200
21
is accepted at the 0.01 level. The difference between 203.5 and
TEST OF A POPULATION PROPORTION
 Proportion is a ratio, fraction, or percentage that gives the part of
the population sample having a particular trait of interest. Thus,
sample proportion is computed as

P=

 Proportion is a value between 0 & 1

22
CONT.…
 With the symbol P denoting the population proportion and Po
denoting a particular hypothesized value for the population
proportion,
 The three forms for hypothesis test about population proportion are
as follows:

i) Ho: P  Po ii) Ho: P  Po iii) Ho: P = Po

Ha: P < Po Ha: p > Po Ha: P  Po

 The first two forms are one - tailed tests, whereas the third form is
23
a two - tailed test.
EXAMPLE

A survey of the morning beverage market of Addis Ababa has shown

that the primary breakfast beverage for 17% of the citizens is milk.
Sholla Wetet, a milk producer in the City, believes that the figure is
higher for Addis Ababa. To test this idea, the researcher Contacts a
random sample of 550 Addis Ababa residents and asks which
primary beverage for breakfast and 115 of them are milk is the
primary beverage for breakfast. Test the idea that the milk figure is
higher for Addis Ababa; using 0.05 a level of significance.

24
SOLUTION
 Step 1: State the hypothesis:
Ho: P 0.17
Ha: P > 0.17
 Step 2: State the decision rule:
 This is a one-tailed test, and the table reading of z 0.05 = +
1.645. Reject Ho if sample Z > 1.645
 Step 3: Compute sample z:

 Step 4: Accept or Reject the Ho:

 Sample Z > 1.645 2.44 > 1.645, True=> Reject the Ho
25
 The producer, Sholla Wetet, is ready to conclude that a higher
proportion of Addis-Ababa residents drink milk as the primary
HYPOTHESIS TESTING OF THE DIFFERENCE
BETWEEN TWO MEANS
 It is a hypothesis test of the difference between two
means to compare two sample means,
example, the average lifetimes of two different brands
of bus tires might be compared to see whether there is
any difference in tread wear.
Two different brands of fertilizer might be tested to see
whether one is better than the other for growing plants.
26
CONT.….
Assumptions for the z Test to Determine the Difference between
Two Means
 Both samples are random samples.
 The samples must be independent of each other. That is, there
can be no relationship between the subjects in each sample.
 The standard deviations of both populations must be known, and
if the sample sizes are less than 30, the populations must be normally or
approximately normally distributed.

27
 Formula for the z Test for Comparing Two Means from
Independent Populations

28
 CONT.…
 These tests can also be one-tailed, using the following hypotheses:

 The tests can also be Two-tailed, using the following hypotheses:

Ho:µ1=µ2 Ho:µ1-µ2=0
Ha:µ1≠µ2 Or Ha:µ-1µ2≠0

29
 CONT.…

 The theory behind testing the difference between two means is

based on selecting pairs of samples and comparing the means of
the pairs.
 If The population means need not be known.
 All possible pairs of samples are taken from populations. The
means for each pair of samples are computed and then subtracted,
and the differences are plotted.
 If both populations have the same mean, then most of the
30
differences will be zero or close to zero.
CONT.…..
 The basic steps/format for hypothesis testing using the traditional
method is reviewed here.
 Step 1; State the hypotheses
 Step 2; Find the critical value(s).
 Step 3; Compute the test value.
 Step 4; Make the decision.
 Step 5; summarize the results.

31
EXAMPLE

A survey found that the average hotel room rate in Addis Ababa is
$88.42 and the average room rate in Gondar is $80.61. Assume that
the data were obtained from two samples of 50 hotels each and that
the standard deviations of the populations are $5.62 and $4.83,
respectively. At α = 0.05, can it be concluded that there is a
significant difference in the rates?

32
SOLUTION
Step 1; State the hypotheses and identify the claim
H0: µ1= µ2
H1: µ1 ≠ µ2 (claim)
Step 2; Find the critical values. Since α = 0.05, the critical values are
+1.96 and -1.96.
Step 3; Compute the test value.

Step 4; Make the decision. Reject the null hypothesis at α = 0.05,

since 7.45 > 1.96. in other word the test stat is greater than the 33
critical value, so we going to reject H0.
See the next Figure.
 CONT.….

Step 5; summarize the results. There is enough evidence to support

the claim that the means are not equal. Hence, there is a significant
difference in the rates.

34
HYPOTHESIS TESTING OF THE DIFFERENCE
BETWEEN TWO PROPORTIONS
 It is used to test the equality of two population proportions.
 For example, a researcher might ask, is the proportion of men who
exercise regularly less than the proportion of women who exercise
regularly? Is there a difference in the percentage of students who
own a personal computer and the percentage of nonstudents who
own one? Is there a difference in the proportion of college
graduates who pay cash for purchases and the proportion of non-
college graduates who pay cash?
35
CONT.…

The formula for the test value is shown next.

Assumptions for the z Test for Two Proportions

 The samples must be random samples.
 The sample data are independent of one another.
 For both samples np ≥ 5 and nq ≥ 5.
36
 EXAMPLE

 A survey of 1000 drivers this year showed that 29% of the people
send text messages while driving. Last year a survey of 1000
drivers showed that 17% of those send text messages while
driving. At α=0.01, can it be concluded that there has been an
increase in the number of drivers who text while driving?

37
SOLUTION

38
CONT.….

39
CONT.….

40
THE END
THANK
YOU
41