Probability and Statistics Notes for Chapter 7: Hypothesis Testing with One Sample

Section 7.1: Introduction to Hypothesis Testing

Hypothesis Test: a process that tests a claim about the value of a population parameter.

Statistical hypothesis: a claim about a population parameter.

To test a statistical hypothesis, a pair of hypotheses must be stated: one that represents the
claim and the other is its complement. Of these two hypotheses, the one that contains a
statement of equality is the null hypothesis. The complement of the null hypothesis is an
alternative hypothesis. Either hypothesis- the null or the alternative may represent the
original claim.

Null Hypothesis: is a statistical hypothesis test that contains a statement of equality,

such as .

Alternative hypothesis: is the complement of the null hypothesis. It is a statement

that contains a statement of inequality, such as <, , >.

History: Ronald Fisher (1890-1962) was the first to identify the null and alternative
hypothesis.

No matter, which hypothesis represents the claim, you always begin a hypothesis test by
assuming that the equality condition in the null hypothesis is true. So, when you perform
a hypothesis test, you make one of two decisions.
1. reject the null hypothesis or
2. fail to reject the null hypothesis.

Hypothesis-Testing Common Phrases:

> <

Is greater than Is less than Is great than or equal to

Is above Is below It at least
Is higher than Is lower than Is not less than
Is longer than Is shorter than
Is bigger than Is smaller than
Is increased Is decreased or reduced from

=
Is
Is equal to Is not equal to
Is less than or equal to Is exactly the same as Is different from
Is at most Has not changed from Has changed from
Is not more than Is the same as Is not the same as

1
Making mistakes in the Justice System: If someone is arrested, they are innocent until
proven guilty. How do some criminals get to go free on technicalities? There are errors.
The alternative hypothesis- the reason a criminal is arrested.
The null hypothesis- presumption of innocence.
Reasonable Doubt- null hypothesis has to be reject beyond reasonable doubt, standard is
often set at 5%- called the alpha level.
Therefore, the legal justice system and statisticians try to disprove or reject the null
hypothesis instead of proving the alternative.

Type I error: reject , when it is actually true, false negative- test says you do not
have the disease, when you really do. There is enough evidence to support the claim.
Example: innocent person goes to jail when they are not guilty (travesty).

Type II error: do not reject , when it is actually false, false positive- test says you do
have the disease, when you really don’t. There is not enough evidence to reject the
claim. Example: guilty person does not go to jail- OJ Simpson or Michael Jackson case.

4 possible outcomes of a hypothesis test:

Decision is true is false

Do not reject 1. Correct decision 2. Type II error

Reject 3. Type I error 4. Correct decision

Defendant is not guilty. Defendant is guilty.

Not guilty verdict 1. Justice 2. Type II error-OJ Simpson
Guilty verdict 3. Type I error 4. Justice

This is sometimes compared to the legal system. The defendant is assumed innocent
until proven guilty. If the evidence is not strong enough, there is no conviction. A “not
guilty” verdict does not prove that a defendant is innocent. The legal system assumes
more harm is done by convicting the innocent (type I error) than by not convicting the
guilty (type II error).

Because there is variation from sample to sample, there is always a possibility you will
reject a null hypothesis when it is actually true. You can decrease the probability of doing
so by lowering the level of significance.

In a hypothesis test level of significance is your maximum allowable probability of

making a type I error. It is denoted by , the lowercase Greek letter alpha.
The probability of a type II error is denoted by , the lowercase Greek letter beta.

One-tailed test: indicates that the null hypothesis should be rejected when the test value
is in the critical region on one side of the mean. A one-tailed test is either right-tailed or
left-tailed, depending on the direction of the inequality of the alternative hypothesis.

2
Two-tailed test: the null hypothesis should be rejected when the test value is in either of
the two critical regions.

Look at the alternative hypothesis to determine which type of test to use.

Left-Tailed Test Right-Tailed Test Two-Tailed Test

Solving Hypothesis-Testing Problems (traditional method):

1. State the hypotheses and identify the claim.
2. Find the critical value(s) from table 4, in the back of your textbook.
3. Compute the test value.
4. Make the decision to reject or not reject the null hypothesis.
5. Summarize the results.

7.2 Hypothesis Testing for the Mean (Large Samples)

P-value (probability value): is the probability of getting a sample statistic (such as the
mean) or a more extreme sample statistic in the direction of the alternative hypothesis
when the null hypothesis is true. P is the area to the left or right of z.

To use a P-value to make a conclusion in a hypothesis test, compare the P-value to .

1. If P , then reject = there is enough evidence …
2. If P > , then fail to reject = there is not enough evidence to …

Is the P-value less than or equal to the level of significance?

Yes- Reject the null hypothesis
No- Fail to reject the null hypothesis

If a P-value is less than 0.01, the null hypothesis will be rejected at the common levels of
= 0.01, = 0.05, = 0.10.
If the P-value is greater than 0.10, then you would fail to reject for these common
levels.
The lower the P-value, the more evidence there is in favor of rejecting . The P-value
gives you the lowest level of significance for which the sample statistic allows you to
reject the null hypothesis.

Finding the P-Value for a hypothesis Test: When given z value and alpha level:
a. For a left-tailed test, P = Area in the left tail
b. For a right-tailed test, P=1-p
c. For a two-tailed test z positive, P = 2(1 - p): Find area, 1 - area, then multiply by 2
d. For a two-tailed test z negative, P = 2 p: Find area, then multiply by 2

z-test: is a statistical test for a population mean. The z-test can be used when the
population is normal and is known, or for any population when the sample size n is at
least 30. The test statistic is the sample mean and the standardized test statistic is z.

3
If n 30, the textbook uses s is used for sample standard deviation.

Calculator: Hypothesis Test for the z Distribution (Statistics)

For finding z and p values, when given the following information:
STAT, TESTS, 1, STAT, ENTER, type values for , move cursor to appropriate
alternative hypothesis, ENTER, CALCULATE, ENTER

Solving Hypothesis-Testing Problems (P-Value Method):

1. State the hypothesis and identify the claim.
2. Compute the test value ( ).
3. Find the P-value (area).
4. Make the decision. Reject of fail to reject the null hypothesis.
5. Summarize the results. Write a statement to interpret the decision in context
of the original claim.

Rejection region (critical region): range of values for which the null hypothesis is
not probable. If a test statistic falls in this region, the null hypothesis is rejected.
Critical value (z) separates the rejection region from the nonrejection region.

Finding Critical Values, : If given alpha and type of test: Use Table 4 in the back of the
textbook, if you cannot find the exact area, use the area that is closest.
Right-tailed test: + answer
Left-tailed test: - answer
Two-tailed test: (take half of alpha), find the value and your answer will have
Commonly Used Levels of Significance
Alpha, Tail z
Left -1.28
0.10 Right 1.28
Two 1.645

Left -1.645
0.05 Right 1.645
Two 1.96

Left -2.33
0.01 Right 2.33
Two 2.575
will never equal 0, because no z-values correspond to = 0. If = 0, the null
hypothesis cannot be rejected and hypothesis test is useless.

*We are skipping Section 7.3- Hypothesis Testing for the Mean (Small Samples)

Section 7.4: Hypothesis Test for Proportion

4
Given a binomial distribution such that np 5 and nq 5, you can use the z-test to test a
proportion p. The test statistic is the sample proportion and the standardized test
statistic is z.
Calculator: Hypothesis Test for the Proportion: STAT, TESTS, 5, type in appropriate
values for , move cursor to the appropriate values for alternative hypothesis,
ENTER, CALCUALTE, ENTER

Using a z-test for a proportion p:

1. State the claim mathematically and verbally. Identify the null and alternative
hypothesis.
2. Specify the level of significance.
3. Determine any critical values.
4. Determine any rejection regions.
5. Find the standardized test statistic, z, calculator or chart.
6. Make a decision to reject or fail to reject the null hypothesis.
7. Interpret the decision in the context of the original claim.