Mr.

Mohamed El-Sayed El-Dawoody Lecturer of Mathematical Statistics

LECTURE 4

Hypothesis Testing

Introduction
 In most fields, the researchers are interested in answering many types of questions and need
to make decisions about these questions. For example ,
 A scientist might want to know whether there is an evidence of global warming or not.
 A physician wants to know whether a new medication will lower a person’s blood pressure.
 An educator wants to see whether a new teaching technique is better than a traditional one.
 A fashion expert might want to know whether the public prefers a certain type of colors in
a new line of fashion.
 An auto specialist is interested in determining whether a new type of seat belt will reduce
the severity of injuries caused by accidents.
 These types of questions can be answered (studied) through a new statistical technique called
"Statistical Hypothesis Testing".
 Hypothesis Testing is a decision-making process interested in testing or evaluating some
claims (hypotheses) about the population parameters.
 In hypothesis testing, the researcher must define each of the following:
 The population under study.
 The data of a sample taken from the population.
 The particular hypotheses that will be investigated.
 The level of significance at which we reject one of the hypotheses.
 Perform the calculations required for the statistical test based on some rules.
 Reach a conclusion that enables us to make the correct decision to solve the problem.
 In this lecture, we will present some basic concepts related to the hypothesis-testing procedure
such as "Statistical Hypothesis, Statistical Test, and Critical Values".
 Also, we will study and explain the hypothesis-testing procedure for testing a claim about the
population mean (µ) using a statistical test is named "z test for the mean".
 Different problems and applications will be mentioned, in this lecture, to cover all topics.

LECTURE 4 PAGE 1 STAT 2040

Mr. Mohamed El-Sayed El-Dawoody Lecturer of Mathematical Statistics

Statistical Hypothesis
 Statistical Hypothesis is a claim (conjecture) about the population parameter and it may
or may not be true.
 There are two basic types of statistical hypothesis are:
1. Null Hypothesis  It is denoted by H0 and is defined as a statistical hypothesis that
states that there is no difference between a population parameter
and a specific value.
2. Alternative Hypothesis  It is denoted by H1 and is defined as a statistical hypothesis
that states that there is a difference between a population
parameter and a specific value.
 To state the null and alternative hypotheses, we have three different cases are summarized as:
DIFFERENT CASES HYPOTHESES TEST NAME WITH WORDS
H0 :  k . Less than, Lower than,
CASE I Left-tailed test Shorter than, Smaller than,
H1 :   k . Is below, Is decreased
H0 :  k . More than, Higher than
CASE II Right-tailed test Longer than, Bigger than
H1 :   k . Is above, Is increased
H0 :  k . Is not equal to, Is different
CASE III Two-tailed test from, Has changed from
H1 :   k . Is not the same as
 The claim (conjecture) can be stated as either H0 or H1; but the statistical evidence can only
support the claim if it is the alternative hypothesis. That is, the statistical evidences can be
used to reject the claim if the claim is the null hypothesis.
Example 1

State the null and alternative hypotheses for each of the following:
(a) A researcher thinks that if expectant mothers use vitamin pills, the birth weight of the babies
will increase. The average birth weight of the population is 8.6 pounds.
(b) A psychologist feels that playing soft music during a test will change the results of the test. The
psychologist is not sure whether the grades will be higher or lower. In the past, the mean of
the scores was 73.
(c) An engineer claims that the mean number of defects can be decreased in a manufacturing
process of compact disks by using robots instead of humans. The mean number of defective
disks per 1000 is 18.
Solution

LECTURE 4 PAGE 2 STAT 2040

Mr. Mohamed El-Sayed El-Dawoody Lecturer of Mathematical Statistics

(a) The null and alternative hypotheses are

H 0 :   8.6 and H 1 :   8.6 (claim).
(b) The null and alternative hypotheses are
H 0 :   73 and H 1 :   73 (claim).
(c) The null and alternative hypotheses are
H 0 :   18 and H 1 :   18 (claim).
Statistical Test
 Statistical test is a test that uses the sample data taken from a certain population to make a
decision about whether the null hypothesis (H0) is rejected or not.
 In the statistical test, there are two possible cases about the null hypothesis are:
 H0 is true  That is, the difference between the population parameter and a specific value,
if exists, is due to chance; or
 H0 is false  That is, the difference between the population parameter and a specific value
is not due to chance, but due to the evidence (significant difference).
 If the difference between the population parameter and a specific value is significant, the
null hypothesis is rejected. If it is not significant, then the null hypothesis is not rejected.
 Also, in the statistical test, there are four possible outcomes, two outcomes for a correct
decision and two outcomes for an incorrect decision. These outcomes are summarized as:
(1) Rejecting the null hypothesis when it is true. This is
an incorrect decision and is called "type I error".
(2) Not rejecting the null hypothesis when it is false. This
is an incorrect decision and is called "type II error".
(3) Rejecting the null hypothesis when it is false. This
would be a "correct decision".
(4) Not rejecting the null hypothesis when it is true. This
would be a "correct decision".
 The level of significance is defined as the maximum probability of occurrence a type I
error and it is denoted by  . That is,
P (type I error)   .
 The most common three levels of significance that will be used are 0.05, 0.01, and 0.10. For
instance, if   0.10 , then there is a 10% chance of rejecting a true null hypothesis.

LECTURE 4 PAGE 3 STAT 2040

Mr. Mohamed El-Sayed El-Dawoody Lecturer of Mathematical Statistics

 The test value is a numerical value obtained from the sample data according to the statistical
test used.
 The critical (rejection) region is the region in which the null hypothesis (H0) is rejected
when the test value lies on it.
 The non-critical (non-rejection) region is the region in which the null hypothesis (H0) is
not rejected when the test value lies on it.
 The critical value is the value that separates the critical region from the noncritical region.
It is denoted by C.V.
 There are two different types of statistical tests are described as:
 One-tailed test  Is a statistical test in which the critical region of a distribution is one-
sided, either left (left-tailed test) or right (right-tailed test).
 Two-tailed test  Is a statistical test in which the critical region of a distribution is two-
sided, one to left and one to right.
Determining the Critical Values
 To obtain the critical value (C.V), in case of the z distribution, we must know the level of
significance  and the type of the statistical test (left, right, or two-tailed).
 The critical values can be computed, in three different cases, using the following rules:
(1) For a left-tailed test
 The hypotheses are
H 0 :   k and H 1 :   k .
 The critical value is given by
C.V  z  .
(2) For a right-tailed test
 The hypotheses are
H 0 :   k and H 1 :   k .
 The critical value is given by
C.V  z 1 .
(3) For a two-tailed test
 The hypotheses are
H 0 :   k and H 1 :   k .
 The critical values are given by
C.V1  z  2 and C.V2  z 1 2 .

LECTURE 4 PAGE 4 STAT 2040

Mr. Mohamed El-Sayed El-Dawoody Lecturer of Mathematical Statistics

Example 2

Use the z distribution tables to find the critical values and draw the appropriate area for each of:
(a) A left-tailed test with
(b) A right -tailed test with
(c) A two-tailed test with
Solution

(a) Draw the appropriate area as follows:

Then, the required critical value is given by

C.V  z   z 0.1000  z 0.1003  1.28.
(b) Draw the appropriate area as follows:

Then, the required critical value is given by

C.V  z 1  z 10.0050  z 0.9950  z 0.9949 or z 0.9951
2.57  2.58 5.15
   2.575  2.58.
2 2
(c) Draw the appropriate area as follows:

Then, the required critical values are given by

C.V1  z  2  z 0.02 2  z 0.0100  z 0.0099  2.33.
C.V2  z 1 2  z 10.0100  z 0.9900  z 0.9901  2.33.

LECTURE 4 PAGE 5 STAT 2040

Mr. Mohamed El-Sayed El-Dawoody Lecturer of Mathematical Statistics

1. z Test for the Mean

 In this lecture, we will study and explain, in more details, the first statistical test used in
hypothesis testing which is called "z test for the mean".
 The z test for the mean is used for testing any claim about the population mean (µ) when
the population standard deviation (σ) is known.
 To test any claim (hypothesis) about the population mean (µ) when σ is known, we must
apply the following steps:
(1) State the null and alternative hypotheses and identify the claim.
(2) Find the critical value(s) using the z distribution tables.
(3) Compute the test value using the formula
x 
z  ,
 n
where,
 n  The sample size.
 x  The sample mean.
   The hypothesized population mean (From H0).
   The population standard deviation (assumed).
(4) Make the decision to reject or not reject the null hypothesis H0.
(5) Summarize the results obtained (reject or support the claim).
Remark
Three assumptions must be satisfied to apply the z test for the mean are:
 The samples must be randomly selected.
 The sample size is greater than or equal to 30 (i.e, population is normally distributed).
 If the sample size is less than 30, the population must be approximately normally distributed.
Example 3

A researcher wishes to see if the mean number of days that a basic, low-price, small automobile sits
on a dealer’s lot is 29 days or not. A sample of 30 automobile dealers has a mean of 30.1 days for
basic, low-price, small automobiles. At a   0.05 , test the claim that the mean time is greater than
29 days? The standard deviation of the population is 3.8 days.
Solution

Since, we have
n  30, x  30.1,   3.8,   0.05.

LECTURE 4 PAGE 6 STAT 2040

Mr. Mohamed El-Sayed El-Dawoody Lecturer of Mathematical Statistics

Then, we get
 The hypotheses are given by
H 0 :   29 and H 1 :   29 (claim).
 The critical value is given by
C.V  z 1  z 10.0500
 z 0.9500  z 0.9495 or z 0.9505
1.64  1.65 3.29
 
2 2
 1.645  1.65.
 The test value is given by
x   30.1  29
z    1.59.
 n 3.8 / 30
 The decision is given by
z  z 1  H0 is not rejected.
 Thus, there is no evidence to support the claim that the mean time is greater than 29 days.
Example 4

A researcher claims that the average cost of men’s athletic shoes is less than $80. He selects a random
sample of 36 pairs of shoes from catalog and finds these costs (rounded to the nearest dollar).
60 70 75 55 80 55 50 40 80 70 50 95
120 90 75 85 80 60 110 65 80 85 85 45
75 60 90 90 60 95 110 85 45 90 70 70
Is there enough evidence to support the researcher’s claim at   0.10 ? Assume   $19.2.
Solution

Since, we have
n  36, x 
x i

60  70  ....  70 2700
  75,   19.2,   0.10.
n 36 36
Then, we get
 The hypotheses are given by
H 0 :   80 and H 1 :   80 (claim).
 The critical value is given by
C.V  z   z 0.1000
 z 0.1003  1.28.

LECTURE 4 PAGE 7 STAT 2040

Mr. Mohamed El-Sayed El-Dawoody Lecturer of Mathematical Statistics

 The test value is given by

x  75  80
z    1.56.
 n 19.2 / 36
 The decision is given by
z  z   H0 is rejected.
 Thus, there is enough evidence to support the claim that the average cost of men’s athletic
shoes is less than $80.
Example 5

The Medical Rehabilitation Education Foundation reports that the average cost of rehabilitation for
stroke victims is $24,672. To see if the average cost of rehabilitation is different at a particular
hospital, a researcher selects a random sample of 35 stroke victims at the hospital and finds that
the average cost of their rehabilitation is $26,343. At   0.01 , can it be concluded that the average
cost of stroke rehabilitation at a particular hospital is different from $24,672? Assume   $3251.
Solution

Since, we have
n  35, x  26,343.1,   3251,   0.01.
Then, we get
 The hypotheses are given by
H 0 :   24,672 and H 1 :   24,672 (claim).
 The critical values are given by
C.V1  z  2  z 0.01 2  z 0.0050
2.57  2.58 5.15
   2.575  2.58.
2 2
C.V2  z 1 2  z 10.0050  z 0.9950
2.57  2.58 5.15
   2.575  2.58.
2 2
 The test value is given by
x   26,343  24, 672
z    3.04.
 n 3251/ 35
 The decision is given by
z  z 1 /2  H0 is rejected.
 Thus, there is enough evidence to support the claim that the average cost of rehabilitation at
the particular hospital is different from $24,672.
‫تمت بـحمـد اللـه‬

LECTURE 4 PAGE 8 STAT 2040