Section 9.

1: Basic Principles of Hypothesis

Testing
OBJECTIVES

1. Define the null and alternate hypotheses

2. Describe the reasoning used in hypothesis testing
3. State conclusions to hypothesis tests
4. Distinguish between Type I and Type II errors

THE INVESTIGATIVE PROCESS OF STATISTICS

We have described statistical analysis as a four-step process whose steps are:

• Formulate questions.
• Collect data needed to answer the questions.
• Describe the data.
• Draw conclusions, using appropriate methods.
We now discuss how to perform hypothesis tests, which is a one of the most commonly used methods
for drawing conclusions.

OBJECTIVE 1
DEFINE THE NULL AND ALTERNATE HYPOTHESES

A study published in the Journal of the Air and Waste Management Association reported that the mean
amount of particulate matter (PM) produced by cars and light trucks in an urban setting is 35 milligrams
of PM per mile of travel. Suppose that a new engine design is proposed that is intended to reduce the
amount of PM in the air. There are two possible outcomes that could happen with the new engine
design: either the new design will reduce the level of PM, or it will not.

These possibilities are called hypotheses. One of the hypotheses is called the null hypothesis and the
other is called the alternate hypothesis.

a
1
Section 9.1: Basic Principles of Hypothesis
Testing
The states that the parameter is equal to a specific value, μ0 . The null
hypothesis is denoted H 0.
The is denoted H 1. There are three possible alternate hypotheses:

Example: State an appropriate null and alternate hypotheses:

Boxes of a certain kind of cereal are labeled as containing 20 ounces. An inspector thinks
that the mean weight may be less than this.

Last year, the mean monthly rent for an apartment in a certain city was $800. A real
estate agent believes that the mean rent is higher this year.

Scores on a standardized test have a mean of 70. Modifications are made to the test,
and an educator believes that the mean may have changed.

2
Section 9.1: Basic Principles of Hypothesis
Testing

CHECK YOUR UNDERSTANDING

State an appropriate null and alternate hypotheses.
a) Last year, the mean amount spent by customers at a certain restaurant was $35. The restaurant
owner believes that the mean may be higher this year.
b) In a recent year, the mean weight of newborn boys in a certain country was 6.6 pounds. A
doctor wants to know whether the mean weight of newborn girls differs from this.
c) A certain model of car can be ordered with either a large or small engine. The mean number of
miles per gallon for cars with a small engine is 25.5. An automotive engineer thinks that the
mean for cars with the larger engine will be less than this.

PURPOSE OF A HYPOTHESIS TEST

The purpose of a hypothesis test is to determine

The idea behind a hypothesis test is similar to a criminal trial. At the beginning of a trial, the defendant is
assumed to be innocent. Then the evidence is presented. If the evidence strongly indicates that the
defendant is guilty, we abandon the assumption of innocence and conclude the defendant is guilty. In a
hypothesis test, the null hypothesis is like the defendant in a criminal trial.

At the start of a hypothesis test, we assume that the null hypothesis is true. Then we look at the
evidence, which comes from data that have been collected. If the data strongly indicate that the null
hypothesis is false, we abandon our assumption that it is true and believe the alternate hypothesis
instead. This is referred to as rejecting the null hypothesis.

3
Section 9.1: Basic Principles of Hypothesis
Testing
OBJECTIVE 2
DESCRIBE THE REASONING USED IN HYPOTHESIS TESTING

EVIDENCE AGAINST THE NULL HYPOTHESIS

We reject the null hypothesis when the data provide strong evidence against it. The evidence comes in
the form of a test statistic. When the difference between the test statistic and the value in the null
hypothesis is sufficiently large, we reject H 0.

Suppose an exam has been given to a large class and the instructor claims that the mean score on the
exam is 80. This is the null hypothesis: . Now suppose that a sample of exam scores is
selected and that the sample mean is . Because 78 is fairly close to 80, it is plausible that the null
hypothesis is true.

Imagine, however, that the sample mean had been . Because 50 is far from 80, it does not seem
plausible that the population mean is actually 80. We would probably reject H 0 in this case.

We see that the farther the test statistic is from the value specified by H 0, the less likely the difference
is due to chance – and the less plausible H 0becomes. The question then is:

To answer this question, we need methods that enable us to calculate just how plausible H 0is.

Hypothesis tests provide these methods taking into account things such as the

and the

4
Section 9.1: Basic Principles of Hypothesis
Testing
OBJECTIVE 3
STATE CONCLUSIONS TO HYPOTHESIS TESTS

If the null hypothesis is rejected, the conclusion is straightforward: We conclude that the alternate
hypothesis, H 1, is true.

If the null hypothesis is not rejected, we say that there is not enough evidence to conclude that the
alternate hypothesis, H 1, is true. This is not saying the null hypothesis is true. What we are saying is that
the null hypothesis might be true.

Example 1: Boxes of a certain kind of cereal are labeled as containing 20 ounces. An inspector thinks
that the mean weight may be less than this, so he performs a test of versus
. State an appropriate conclusion if the inspector rejects H 0.

Solution:

Example 2: Consider the previous example, state an appropriate conclusion if the inspector does
not reject H 0.

Solution:

5
Section 9.1: Basic Principles of Hypothesis
Testing

OBJECTIVE 4
DISTINGUISH BETWEEN TYPE I AND TYPE II ERRORS

When a hypothesis test is conducted and a decision is made there is a possibility that it is the wrong
decision.

There are two ways in which a wrong decision may occur with hypothesis testing:

1.

2.

In general, a Type I error is more serious than a Type II error. This is because a Type I error results in a
false conclusion, while a Type II error results only in no conclusion.

Ideally, we would like to minimize the probability of both errors. Unfortunately, with a fixed sample size,
decreasing the probability of one type increases the probability of the other.

Hypothesis tests are often designed so that the probability of a Type I error will be acceptably small,

often 0.05 or 0.01. This value is called the of the test.

6
Section 9.1: Basic Principles of Hypothesis
Testing
Example: The dean of a business school wants to determine whether the mean starting salary of
graduates of her school is greater than $50,000. She will perform a hypothesis test with
the following null and alternate hypotheses:

Decide if a Type I error, Type II error, or a correct decision occurs in each case:

a) The true mean is μ=$ 50,000 , and the dean rejects H 0.

Solution:

b) The true mean is μ=$ 55,000 , and the dean rejects H 0.

Solution:

c) The true mean is μ=$ 55,000 , and the dean does not reject H 0.

Solution:

CHECK YOUR UNDERSTANDING

Decide if a Type I error, Type II error, or a correct decision occurs in each case:
a) A test is made of versus . The true value of µ is 150 and H 0 is
rejected.
b) A test is made of versus . The true value of µ is 20 and H 0 is not
rejected.
c) A test is made of versus . The true value of µ is 3 and H 0 is rejected.

7
Section 9.1: Basic Principles of Hypothesis
Testing
YOU SHOULD KNOW . . .

 How to write the null and alternate hypotheses

 How to determine whether a hypothesis test is left-tailed, right-tailed, or two-tailed

 How to state the conclusion to a hypothesis test

 How to distinguish between Type I and Type II errors and correct decisions