You are on page 1of 17

Chapter 10.

Hypothesis Testing Using a Single Sample


Basics of hypothesis testing
In this section, 1st we introduce the language of hypothesis testing, then we discuss the
formal process of testing a hypothesis.
Definitions:

A hypothesis is a statement or claim regarding a characteristic of one or

more populations.
Hypothesis testing (or test of significance) is a procedure, based on a
sample evidence and probability, used to test claims regarding a characteristic of one or
more populations.
To test a hypothesis, you should carefully state a pair of hypothesis one that represents
the claim and the other, its complement.
Goals:

How to identify the null hypothesis and alternative hypothesis from a given claim,
and how to express both in symbolic form

How to calculate the value of the test statistic, given a claim and sample data

How to choose the sampling distribution that is relevant

How to identify the P-value or identify the critical value(s)

How to state the conclusion about a claim in simple and nontechnical terms

Definitions:

Null Hypothesis (Ho) the hypothesis regarding the population

parameter that is assumed to be true (status quo).


Alternative Hypothesis (Ha) the hypothesis that is counter to the null
hypothesis. Usually this is the hypothesis regarding the population parameter that the
researcher wants to test.
Rule of rare events: If, under a given assumption, the probability of a particular
observed event is exceptionally small, we conclude that the assumption is probably
______ correct.

If carrying out a test of Ho vs. Ha, the null hypothesis Ho will be rejected in favor of Ha
only if sample evidence strongly suggests that Ho is false. If the sample does not provide
such evidence, Ho will not be rejected.
The two possible conclusions of the hypothesis testing procedure are: reject Ho or fail
to reject Ho.
Null hypothesis is:

Ho: population characteristic = hypothesized value

Alternative hypothesis is:

Ha: population characteristic > hypothesized value


Ha: population characteristic < hypothesized value
Ha: population characteristic

hypothesized value

Example 1: Write the claim as a mathematical sentence. State the null and alternative
hypotheses, and identify which represents the claim.
a) A water faucet manufacturer claims that the mean flow rate of a certain type of faucet
is less than 2.5 gallons per minute.
b) A cereal company claims that the mean weight of the contents of its 20-ounce size
cereal boxes is more than 20 ounces.
c) The standard deviation of IQ scores of actors is equal to 15.

Large-Sample Hypothesis Tests for a Population Proportion.


Example 2: The authors of the article Perceived Risks of Heart Disease and Cancer
Among Cigarette Smokers expressed the concern that a majority of smokers do not view
themselves as being at increased risk of heart disease or cancer. A study of 737 current
smokers selected at random from U.S. households found that of the 737 smokers
surveyed, 295 indicated that they believed that they have a higher than average risk of
cancer. Do this data suggest that p the true proportion of smokers who view themselves as
being at increased risk of cancer is less than half, as claimed by the authors of the paper?
Population
Population parameter
Sample
Sample proportion:
Idea: where on the sampling distribution is the sample proportion value located? How far
from the true mean (p)? We reject Ho when the sample we observed gave the statistic
value too far from the hypothesized center.
is away from the assumed center most likely ____________ center
If p
is close to the assumed center most likely________ center
If p

So, the sample gave us the value of sample proportion that is _______ standard
deviations ________ the hypothesized mean of ______.

That means if null hypothesis is true and the true proportion is _______, then there is a
very _____ chance that we got the sample from that distribution. That itself means that
the true proportion must be _________________.
So we would ________ the null hypothesis in favor of alternative hypothesis.
Definitions:

Test Statistic Statistics computed fro the sample that you test with Z

or t.
Rejection Region values of the test statistic that lead to he rejection of
null hypothesis.
P-value the probability of observing a value of the test statistic that is at
least as contradictory to the Null Hypothesis as the one obtained from the sample.
Level of significance () is the probability of making the Type I error
(rejecting the Null Hypothesis when it is indeed true). In other words this is the
percentage of all sample statistics under the null distribution that you are willing to say
that they are unusual.

Go back to example 2:

Let 0.01 , meaning that we will only reject null hypothesis if the sample gave us the
observed sample statistics that is the 1% of most unusual observations under the null
distribution.
P value - reject Ho meaning that this data does suggest that p the true proportion

of smokers who view themselves as being at increased risk of cancer is less than half, as
claimed by the authors of the paper

Decision Rule Based on Rejection (Critical) Region:


To use a rejection region to conduct a hypothesis test, calculate the standardized test
statistic (z or t). If the standardized test statistic
1. Is in the rejection region, then reject H0.
2. Is not in the rejection region, then fail to reject H0.
If we fail to reject the null hypothesis, it does not mean that you have accepted the null
hypothesis as true. It simply means that there is not enough evidence to reject the null
hypothesis.
We cannot actually conclude that we accept the null hypothesis, since we were not
actually provided with null hypothesis, but rather created it ourselves based on the
alternative hypothesis. That is why we say fail to reject null rather than accept null

Errors in Hypothesis Testing


Definitions:

Type I Error rejecting Null Hypothesis when it is actually true

Type II Error failing to reject Null Hypothesis when it is actually false

NOTE: With the hypothesis we are never just right and wrong. Different types of errors
can have dramatic impact.
The choice of the level of significance depends on the consequences of making type I
error. If the consequences are severe, the level of significance should be small.
Which error would you rather make?
Reject a perfectly good parachute and refusing to jump
Fail to reject a defective parachute and jump with it
Go back to example 2:
We reached the conclusion that the null hypothesis should be rejected. What kind of error
can we be making? Explain the error in terms of the problem, find its probability.

Example 3: Provide statements explaining what it would mean to make a type I error and
type II error if a researcher for the FDA whishes to test the claim that the percentage of
children taking the new antibiotic who experience headaches as a side effect is more than

2%. , a) suppose the sample evidence indicates that the null hypothesis is rejected. State
the conclusion. b) Suppose the sample evidence indicated that the null hypothesis is not
rejected. State the conclusion.
Which error has more dramatic consequences?
Which significance level would you choose: 10% or 1% ?

Example 5: Suppose that you are an inspector for the Fishing Department and that you
are given the task of determining whether to prohibit fishing along part of the Oregon
coast. You will close the area to fishing if it is determined that fish in that region have
unacceptably high mercury content.

a. Assuming that a mercury content of 5ppm is considered the maximum safe


concentration, which of the following parts of hypotheses would you test?

H 0 : 5 _ versus _ H a : 5

H 0 : 5 _ versus _ H a : 5

b. Would you prefer a significance level of 0.1 or 0.01 for your test, explain.

Steps in Large-Sample Hypothesis Test for


population proportion p

1. Describe the population and population characteristic that the test will be performed
for
2. Describe the sample.
3. State null and alternative hypothesis, state the significance level

n p 10

4. Check the normality of Sampling Distribution assumptions: n (1 p ) 10


5. Compute test Statistic

p p
p (1 p ) and draw a picture of the sampling
n

distribution.
6. Indicate the P-value on the sampling distribution, compute the P-value
7. Draw a conclusion P value then reject Ho
. P value then fail to reject Ho
reflect on the possible error you could be making.

Example 4. A medical researcher claims that less than 20% of American adults are
allergic to a medication. In a random sample of 100 adults, 15% say they have such an
allergy. Test the researchers claim at significance level of 0.01.

Example 5. The report on education shows that nationwide 61% of high school graduates
go on to attend a two-year of four-year college the year after graduation. The collegegoing rate for high school graduates in California was estimated to be 55%. Suppose that
the estimate was based on a random sample of 1500 California high school graduates in
2009. Can we reasonably conclude that the proportion of California high school
graduates in 2009 who attended college in the year after graduation is different from the
nationwide figure? Test the relevant hypothesis at a 5% significance level.

Example 6. the report 2007 Electronic Monitoring and Surveillance Survey


summarized the results of a survey of 304 U.S. businesses. Of these companies, 201
indicated that they monitor employees web site visits. For purposes of this exercise,
assume that it is reasonable to regard this sample as representative of all U.S. businesses.
Is there sufficient evidence to conclude that more than 60% of U.S. busunesses monitor
employees web site visits. Test appropriate hypothesis using a significance level of 0.01.

10.4 Hypothesis Tess for Population Mean.

I) Population standard
deviation is known
(unrealistic)

II) Population standard


deviation is UNknown
(we will need to estimate
it)

Steps in Hypothesis Test for population mean. ( )


1. Describe the population and population
characteristic that the test will be performed for
2. Describe the sample.
3. State null and alternative hypothesis, state the
significance level

4. Check the normality of Sampling Distribution


assumptions:

Population is normally distributed

Sample size n 30

5. Compute test Statistic and draw a picture of the


sampling distribution.
If is known, use

If is not known, use

x
s , df n 1
n

6. Indicate the P-value on the sampling distribution, compute the P-value


7. Draw a conclusion P value then reject Ho
. P value then fail to reject Ho
reflect on the possible error you could be making.

Example 7. A credit bureau analysis of undergraduate students credit records found that
the average number of credit cards in an undergraduates wallet was 4.09. It was also
reported that in a random sample of 132 undergraduates, the sample mean number of
credit cards carried was 2.6. The sample standard deviation was reported to be 1.2. Is
there convincing evidence that the mean number of credit cards that undergraduates
report carrying is less than a credit bureaus figure of 4.09?

Example 8. Speed, size, and strength are thought to be important factors in football
performance. The article Physical and Performance Characteristics of NCAA Division I
Football Players reported on physical characteristics of Division I starting football
players in 1988 football season. Information for teams ranked in the top 20 was easily
obtained, and it was reported that the mean weight of top 20 teams was 105 kg. A random
sample of 33 starting players from Division I teams that were not ranked in the top 20
resulted in the sample mean weight of 103.3 kg and a sample standard deviation of 16.3
kg.
a. Is there sufficient evidence to conclude that the mean weight for non top starters differs
from value 105, the known value for top 20 teams
b. Is there sufficient evidence to conclude that the mean weight for non top starters if less
than 105, the known value for top 20 teams?

10.5 Power and Probability of Type II Error.


Recall that
Type I Error rejecting Null Hypothesis when it is actually true
Type II Error failing to reject Null Hypothesis when it is actually false

Controlling type I and II errors:

For any fixed , an increase in the sample size n will cause a _______________
in

For any fixed sample size n, a decrease in will cause an __________________


in . Conversely, an increase in will cause a _______________________ in .

To decrease both and , ____________________ the sample size.

Definition: The power of a hypothesis test is the probability 1 of rejecting a false


null hypothesis.
In other words, when Null is false, the power of the test is 1 .
Because if the null is false, you want to indicate so with the hypothesis test, by rejecting
it. The smaller the Type II error (fail to reject false null), the larger the 1

- power.

The value of the power is computed by using a particular significance level and a
particular value of the population parameter that is an alternative to the value assumed
true in the null hypothesis.

You might also like