SOCI1005

Hypothesis Testing

Friday, April 9, 2021

 2

What is hypothesis testing?

• As in interval estimation, the central problem is to make inferences about
a population parameter based upon sample information.

• Hypothesis testing is a procedure which allows us to test whether or not a

parameter is equal to a specified value base on a preconceived notion.
Simply put, hypothesis testing examines a claim or belief or guess about
the value of one or more population parameters.

• The procedure involves comparing the hypothesized value of the

population with the information contained in a sample.

• Hypothesis testing will be performed for a single population mean (μ)

and the difference between two population means (μ1 – μ2).
 Hypothesis Testing
3

• Hypothesis testing is the one of the primary ways we derive inference, since
it provides a technique for us to scientifically and reliably decide if the
evidence supports the theory.

• Hypothesis testing is conducted by following these seven steps:

1. Determine the null hypothesis
2. Determine the alternative hypothesis
3. State the level of significance
4. Determine the rejection region
5. Compute test statistic value
6. Decide whether or not to reject the H0 (by comparing steps 4 and 5).
7. State the conclusion
 4
vA hypothesis is a claim or assumption.

Null hypothesis (H0): It is a claim (or statement) about a

population parameter that is assumed to be true. It is the
value of the population parameter that is taken as being true
before evidence is found against it. In other words, it is what
is generally believed until it is disproved.

Alternative hypothesis (Ha): It is the belief or claim being

investigated by the researcher so as to find evidence in
support of it. This hypothesis determines the type of test to
be performed.
 5

Hypothesis Testing
• Non-directional hypothesis/Two-tailed test: It is a two-tailed test
if the researcher does not take a position on the value of the
parameter. He does not care or know whether the value of the
population parameter (or difference between two parameters) is less
than or greater than a certain value; he just wants sample evidence
to support that it is not equal to that certain value.
• Directional hypothesis/ One-tailed test: This type of test
examines if the population parameter (or difference between two
parameters) is greater than a certain value or if it is less than a
certain value.
Ho Ha

Equal (=) Not equal (≠) or greater than (>)

or less than (<)

Greater than or equal to (≥) Less than (<)

Less than or equal to (≤) More than (>)

 6
• Level of significance ( a ): It is a probability value that defines the size of the
rejection region. This is the probability of rejecting the null hypothesis when it is
true. The significance level is typically, 1%, 5% or 10%. If the level of
significance is not given, assume it to be 5%.

• Rejection (or critical) region: This is the region that allows for the rejection of
the H0. The Z-values that separate the rejection region from the non-rejection
region are called critical values. The position of the rejection region depends on
the type of test.
 7
Critical Values (Z)

Two-tailed Test One-tailed Test

10% 1.645 1.28

5% 1.96 1.645

1% 2.58 2.33
 8

Hypothesis Testing
Test statistic: It is a sample statistic which computes a value that is used to
determine whether to actually reject H0. If this value falls in the rejection region,
reject the H0; otherwise, do not reject H0.
Decision:
q Ifthe test statistics value falls in the region of rejection, we reject the null
hypothesis.
q Ifthe test statistics value falls in the non- rejection region, we fail to reject (or we
accept) the null hypothesis.
There are two possible types of errors that can be made when making a
statistical decision, i.e. accepting or rejecting the null hypothesis (H0).
Type I error: occurs whenever the null hypothesis is rejected and it is true.

Type II error: occurs whenever the null hypothesis is accepted and it is false.
 9

Writing the Null Hypothesis (Ho)

 10
Writing the Alternative Hypothesis (Ha)
• Two-tailed test:
In this case, the alternative hypotheses are written as:

• Ha : μ ≠ c [used for a single population mean, where c is a given value]

• Ha: μ1 – μ2 ≠ 0 [used when finding the difference between two population
means]

• One-tailed test:
In this case, the alternative hypotheses are written as:

Ø For single population mean

• Ha : μ < c [This is a left-tail test]

• Ha : μ > c [This is a right-tail test]

Ø For difference between two population means

• Ha: μ1 – μ2 < 0 / μ1 < μ2 [This is a left-tail test]

• Ha: μ1 – μ2 > 0 / μ1 > μ2 [This is a right-tail test]
 11
State the null and alternative hypotheses for each of the following
situations.
a. The average July temperature in a region historically has been
74.5°F. Perhaps it is lower now.
b. The average stipend for doctoral students in a particular discipline
at a state university is $14,756. The department chairman believes
that the national average is higher.
c. The average room rate in hotels in a certain region is $82.53. A
travel agent believes that the average in a particular resort area is
different.
d. The average time workers spent commuting to work in Verona five
years ago was 38.2 minutes. The Verona Chamber of Commerce
asserts that the average is higher now.
 12
• Test Statistics:

This test statistic is This test statistic is used

used when dealing when dealing with the
with a single mean. difference between two
means.

x1 - x2
x-µ Z=
Z= s 12 s 22
s +
n n1 n2
 13
Worked Example
A random sample of 26 psychology Step 1: μ = 440
majors at a local college scored an Step 2: μ > 440
average of 458 on the Graduate Record Step 3: Level of significance (α) = 5%
Examination verbal skills test, with a Step 4: Critical value: Z = 1.645 (Since it is a
standard deviation of 20. The national one-tailed test at the 5% level of
significance)
average on this test for all psychology
Step 5: Test statistics – Z = σX − µn
majors is 440. Can we conclude that the 458 − 440
Z =
average score at the local college is 20
26
higher than the national average? Test at Z =
20
18
5.10
the 5% significance level. Z =
18
3.92

Solution:
Z = 4.59

Step 6: Since the test statistic value of Z (4.59)

This is a test of a single mean is greater than the critical value of Z (1.645),
and it’s a one-tail test because we reject the null hypothesis (Ho) at the 5%
significance level.
we are testing if the average is
higher. X = 458 Step 7: There is sufficient evidence to indicate
µ = 440 that the average score at the local college is
higher than the national average
σ = 20
 14

Worked Example
The average household size in a
certain region several years ago
was 3.14 persons. A sociologist
wishes to test, at the 5% level of
significance, whether it is
different now. Perform the test
using the information collected
by the sociologist: in a random
sample of 75 households, the
average size was 2.98 persons,
with sample standard deviation
0.82 person.
 Worked Example 15
A training manager in a large industrial
firm claims that employees taking his
course perform better on the job than
those not receiving this training. Of 72
recently hired employees, 36 were
randomly selected to receive training.
The other 36 received no special training.
Six months later, an on-the-job
performance test showed that the trained
group averaged 84.6 with a standard
deviation of 3.5. For the group with no
training, the average was 81.5 with a
standard deviation of 4.0. Test the
manager’s claim at the 1% significance
level.

Solution:

ü It is a one tailed test to the right.

ü It is a test for difference between two

population means.
 Examples 16

1. Blood glucose levels for obese patients have a mean of 100 with a standard deviation of
15. A researcher thinks that a diet high in raw cornstarch will have a positive or negative
effect on blood glucose levels. A sample of 30 patients who have tried the raw cornstarch
diet have a mean glucose level of 140. Test the hypothesis that the raw cornstarch had an
effect.
2. A teachersʼ union would like to establish that the average salary for high school teachers
in a particular state is less than $32,500. A random sample of 100 public high school
teachers in the particular state has a mean salary of $31, 578. It is known from history
that the standard deviation of the salaries for the teachers in the state is $4,415. Test the
unionʼs claim at the 5% level of significance.
3. The average earnings of a sample of 1500 female workers who are not union members is
$909 per week. The average earnings of a sample of 2000 female workers who are union
members is $1035 per week. The standard deviations for the two corresponding
populations are $70 and $90, respectively. Test at a 5% significance level whether the
mean weekly earnings of female workers who are not union members are less than those
of female workers who are union members.
17
 Examples 18

5. A test of breaking strengths of two types of cables, A and B, was conducted

using samples of pieces of each type of cable. The results are summarized in the
table below:
Cable A Cable B
n = 100 n = 100
X = 1925kg X = 1905kg
s = 40 kg s= 30 kg

Is there sufficient evidence to indicate a difference between the mean breaking

strengths of the two cables? Use a significance level of 0.05.