0 оценок0% нашли этот документ полезным (0 голосов)

3 просмотров11 страницJul 13, 2020

© © All Rights Reserved

0 оценок0% нашли этот документ полезным (0 голосов)

3 просмотров11 страницВы находитесь на странице: 1из 11

men than convict one

Inferential Data Analysis-I innocent one.”

-Justice Oliver Wendell Holmes,

Supreme Court of the United States (1882-1899)

Hypothesis Testing

n Developing Null and Alternative Hypotheses

n Type I and Type II Errors

n Population Mean: s Known

n Population Proportion

n Hypothesis Testing and Decision Making

n Calculating the Probability of Type II Errors

n Determining the Sample Size for

a Hypothesis Test About a Population mean

n Hypothesis testing can be used to determine whether • It is not always obvious how the null and alternative

a statement about the value of a population parameter hypotheses should be formulated.

should or should not be rejected.

• Care must be taken to structure the hypotheses

n The null hypothesis, denoted by H 0 , is a tentative appropriately so that the test conclusion provides

assumption about a population parameter. the information the researcher wants.

n The alternative hypothesis, denoted by H a, is the • The context of the situation is very important in

opposite of what is stated in the null hypothesis. determining how the hypotheses should be stated.

n The hypothesis testing procedure uses data from a • In some cases it is easier to identify the alternative

sample to test the two competing statements hypothesis first. In other cases the null is easier.

indicated by H 0 and H a.

• Correct hypothesis formulation will take practice.

1

Developing Null and Alternative Hypotheses Developing Null and Alternative Hypotheses

• Many applications of hypothesis testing involve • Example:

an attempt to gather evidence in support of a A new teaching method is developed that is

research hypothesis. believed to be better than the current method.

• In such cases, it is often best to begin with the • Alternative Hypothesis:

alternative hypothesis and make it the conclusion The new teaching method is better.

that the researcher hopes to support.

• The conclusion that the research hypothesis is true

• Null Hypothesis:

The new method is no better than the old method.

is made if the sample data provide sufficient

evidence to show that the null hypothesis can be

rejected.

Developing Null and Alternative Hypotheses Developing Null and Alternative Hypotheses

• Example: • Example:

A new sales force bonus plan is developed in an A new drug is developed with the goal of lowering

attempt to increase sales. blood pressure more than the existing drug.

• Alternative Hypothesis: • Alternative Hypothesis:

The new bonus plan increase sales. The new drug lowers blood pressure more than

the existing drug.

• Null Hypothesis:

The new bonus plan does not increase sales. • Null Hypothesis:

The new drug does not lower blood pressure more

than the existing drug.

Developing Null and Alternative Hypotheses Developing Null and Alternative Hypotheses

• We might begin with a belief or assumption that • Example:

a statement about the value of a population The label on a soft drink bottle states that it

parameter is true. contains 67.6 fluid ounces.

assumption and determine if there is statistical The label is correct. µ > 67.6 ounces.

evidence to conclude that the assumption is

• Alternative Hypothesis:

incorrect.

The label is incorrect. µ < 67.6 ounces.

• In these situations, it is helpful to develop the null

hypothesis first.

2

Summary of Forms for Null and Alternative Null and Alternative Hypotheses

Hypotheses about a Population Mean

■ The equality part of the hypotheses always appears ■ Example: Metro EMS

in the null hypothesis. A major west coast city provides one of the most

n In general, a hypothesis test about the value of a comprehensive emergency medical services in the

population mean µ must take one of the following world. Operating in a multiple hospital system

three forms (where µ0 is the hypothesized value of with approximately 20 mobile medical units, the

the population mean). service goal is to respond to medical emergencies

with a mean time of 12 minutes or less.

H 0 : µ ³ µ0 H 0 : µ £ µ0 H 0 : µ = µ0 The director of medical services wants to

H a : µ < µ0 H a : µ > µ0 H a : µ ¹ µ0 formulate a hypothesis test that could use a sample

of emergency response times to determine whether

One-tailed One-tailed Two-tailed or not the service goal of 12 minutes or less is being

(lower-tail) (upper-tail) achieved.

The emergency service is meeting n Because hypothesis tests are based on sample data,

H 0: µ < 12 we must allow for the possibility of errors.

the response goal; no follow-up

action is necessary. ■ A Type I error is rejecting H0 when it is true.

■ The probability of making a Type I error when the

The emergency service is not

H a: µ > 12 null hypothesis is true as an equality is called the

meeting the response goal;

level of significance.

appropriate follow-up action is

necessary. ■ Applications of hypothesis testing that only control

the Type I error are often called significance tests.

where: µ = mean response time for the population

of medical emergency requests

Population Condition

■ It is difficult to control for the probability of making

a Type II error. H0 True H0 False

Conclusion (µ < 12) (µ > 12)

■ Statisticians avoid the risk of making a Type II

error by using “do not reject H 0” and not “accept H 0”. Do not reject H0 Correct

Type II Error

(Conclude µ < 12) Decision

Reject H0 Correct

Type I Error Decision

(Conclude µ > 12)

3

Critical Value Approach to

One-Tailed Hypothesis Testing

n The test statistic z has a standard normal probability

distribution.

Inference n We can use the standard normal probability

of a in the lower (or upper) tail of the distribution.

Population Parameter n The value of the test statistic that established the

boundary of the rejection region is called the

critical value for the test.

■ The rejection rule is:

• Lower tail: Reject H0 if z < -za

• Upper tail: Reject H 0 if z > za

Lower-Tailed Test About a Population Mean: Upper-Tailed Test About a Population Mean:

s Known s Known

■ Critical Value Approach ■ Critical Value Approach

Sampling Sampling

distribution distribution

of z =

x - µ0 x - µ0

s/ n of z =

Reject H 0 s/ n Reject H0

a = .10 a = .05

Do Not Reject H 0 Do Not Reject H0

z z

-za = -1.28 0 0 za = 1.645

One-Tailed Hypothesis Testing

n The p-value is the probability, computed using the n Less than .01

test statistic, that measures the support (or lack of Overwhelming evidence to conclude Ha is true.

support) provided by the sample for the null ■ Between .01 and .05

hypothesis. Strong evidence to conclude Ha is true.

n If the p-value is less than or equal to the level of

■ Between .05 and .10

significance a, the value of the test statistic is in the

rejection region. Weak evidence to conclude Ha is true.

Insufficient evidence to conclude Ha is true.

4

Lower-Tailed Test About a Population Mean: Upper-Tailed Test About a Population Mean:

s Known s Known

p-Value Approach p-Value < a , p-Value < a ,

■ ■ p-Value Approach

so reject H0 . so reject H0 .

Sampling

a = .10 Sampling distribution a = .04

x - µ0

distribution of z =

s/ n

x - µ0

of z =

s/ n

p-value p-Value

= .072 = .011

z z

z = -za = 0 0 za = z=

-1.46 -1.28 1.75 2.29

Step 1. Develop the null and alternative hypotheses. Critical Value Approach

Step 2. Specify the level of significance a. Step 4. Use the level of significance to determine the

Step 3. Collect the sample data and compute the critical value and the rejection rule.

value of the test statistic. Step 5. Use the value of the test statistic and the rejection

rule to determine whether to reject H0.

p-Value Approach

Step 4. Use the value of the test statistic to compute the

p-value.

Step 5. Reject H 0 if p-value < a.

One-Tailed Tests About a Population Mean: One-Tailed Tests About a Population Mean:

s Known s Known

■ Example: Metro EMS n p -Value and Critical Value Approaches

The response times for a random sample of 40

medical emergencies were tabulated. The sample 1. Develop the hypotheses. H0: µ < 12

mean is 13.25 minutes. The population standard Ha: µ > 12

deviation is believed to be 3.2 minutes.

The EMS director wants to perform a hypothesis 2. Specify the level of significance. a = .05

test, with a .05 level of significance, to determine

whether the service goal of 12 minutes or less is 3. Compute the value of the test statistic.

being achieved. x - µ 13.25 - 12

z= = = 2.47

s / n 3.2 / 40

5

One-Tailed Tests About a Population Mean: One-Tailed Tests About a Population Mean:

s Known s Known

n p –Value Approach ■ p –Value Approach

distribution a = .05

For z = 2.47, cumulative probability = .9932. x - µ0

of z =

s/ n

p–value = 1 - .9932 = .0068

p-value

5. Determine whether to reject H 0. = .0068

Because p–value = .0068 < a = .05, we reject H 0.

There is sufficient statistical evidence z

to infer that Metro EMS is not meeting 0 za = z=

the response goal of 12 minutes. 1.645 2.47

s Known Does TSM Café needs a Billing Machine

n Critical Value Approach The manager of TSM Café is thinking about establishing a

d 11/22/10 6:31 PM Page 383

new billing system for the store’s credit customers. After

4. Determine the critical value and rejection rule. a thorough financial analysis, he determines that the new

system will be cost-effective only if the mean monthly

For a = .05, z.05 = 1.645

account is more than Rs. 170. A random sample of 400

Reject H 0 if z > 1.645 monthly accounts is drawn, for which the sample mean is

Rs. 178. The manager knows that the accounts are

5. Determine whether to reject H 0. approximately normally distributed with a standard

deviation of Rs 65. Can the manager conclude from this

Because 2.47 > 1.645, we reject H 0.

There is sufficient statistical evidence

thatI Nthe

T Rnew

O D Usystem

C T I O will

N TO be H YPOTHESIS TESTING

cost-effective?

383

to infer that Metro EMS is not meeting

Data

the response goal of 12 minutes.

asked how many minutes of sports he watched on shown here. Assuming that the standard deviation is

television daily. The responses are listed here. It is .05 inch, can we conclude at the 5% significance

known that ! ! 10. Test to determine at the 5% sig- level that the mean diameter is not .50 inch?

nificance level whether there is enough statistical

.48 .50 .49 .52 .53 .48 .49 .47 .46 .51

evidence to infer that the mean amount of television

watched daily by all young adult men is greater than 11.34 Xr11-34 Spam e-mail has become a serious and costly

50 minutes. nuisance. An office manager believes that the aver-

Illustration 2 3

age amount of time spent by office workers reading

50 48 65 74 66 37 45 68 64

andSpam

deleting spam

e-mail has becomeexceeds 25 costly

a serious and minutes per day.

nuisance.

65Federal

58 Express (FedEx) sends invoices to customers requesting

55 52 63 59 57 74 65

payment within 30 days. Each bill lists an address, and customers

are expected to use their own envelopes to return their payments.

ToAn office manager believes that the average amount of

test this belief, he takes a random

time spent by office workers reading and deleting spam sample of

11.30 Xr11-30 Thetheclub

Currently, professional

mean and at ofathedifficult

standard deviation public

amount of time 18 exceeds

workers and measures

25 minutes per day. Tothetest amount

this belief, of time each

he takes

taken to pay bills are 24 days and 6 days, respectively. The chief

course boasts

financial officerthat

(CFO)his course

believes is so tough

that including a stampedthatself- the spends reading

a random sampleand of deleting

18 workersspam. The results

and measures the are

amount of time each spends reading and deleting spam.

average golfer loses a dozen or more golf balls

addressed (SSA) envelope would decrease the amount of time. She

dur- listed here. If the population of times is normal

The results are listed here. If the population of times is with

calculates that the improved cash flow from a 2-day decrease in the

ing apayment

round of would

period golf.payA for

dubious

the costs ofgolfer sets out

the envelopes and to a standard

normal withdeviation

a standardof 12 minutes,

deviation can the

of 12 minutes, can manager

the

manager infer at the 1% significance level that he is

stamps. Any further decrease in the payment period would

showgenerate

that the proToistest

a profit. fibbing. He

her belief, sheasks a random

randomly selects 220sam- infer at the 1% significance level that he is correct?

correct?

ple ofcustomers

15 golfers who just completed their rounds

and includes a stamped self-addressed envelope with

their invoices. The numbers of days until payment is received were

to 35 48 29 44 17 21 32 28 34

report the number

recorded. Can the CFO of golfthat

conclude balls each

the plan lost.

will be Assuming

profitable?

23 13 9 11 30 42 37 43 48

that Data

the number of golf balls lost is normally distrib-

uted with a standard deviation of 3, can we infer at The following exercises require the use of a computer and soft-

the 10% significance level that the average number ware. The answers may be calculated manually. See Appendix A

of golf balls lost is less than 12? for the sample statistics.

1 14 8 15 17 10 12 6 11.35 Xr11-35 A manufacturer of lightbulbs advertises that,

14 21 15 9 11 4 8 on average, its long-life bulb will last more than

5,000 hours. To test the claim, a statistician took a

11.31 Xr11-31 A random sample of 12 second-year university random sample of 100 bulbs and measured the

6

students enrolled in a business statistics course was amount of time until each bulb burned out. If we

drawn. At the course’s completion, each student was assume that the lifetime of this type of bulb has a

asked how many hours he or she spent doing home- standard deviation of 400 hours, can we conclude at

work in statistics. The data are listed here. It is known the 5% significance level that the claim is true?

that the population standard deviation is ! ! 8.0. The

p-Value Approach to Critical Value Approach to

Two-Tailed Hypothesis Testing Two-Tailed Hypothesis Testing

n Compute the p-value using the following three steps: n The critical values will occur in both the lower and

1. Compute the value of the test statistic z. upper tails of the standard normal curve.

2. If z is in the upper tail (z > 0), compute the n Use the standard normal probability distribution

probability that z is greater than or equal to the table to find za/2 (the z-value with an area of a/2 in

the upper tail of the distribution).

value of the test statistic. If z is in the lower tail

(z < 0), compute the probability that z is less than or ■ The rejection rule is:

equal to the value of the test statistic. Reject H0 if z < -za/2 or z > za/2.

3. Double the tail area obtained in step 2 to obtain

the p –value.

n The rejection rule:

Reject H 0 if the p-value < a .

Two-Tailed Tests About a Population Mean: Two-Tailed Tests About a Population Mean:

s Known s Known

■ Example: Glow Toothpaste ■ Example: Glow Toothpaste

The production line for Glow toothpaste is Assume that a sample of 30 toothpaste tubes

designed to fill tubes with a mean weight of 6 oz. provides a sample mean of 6.1 oz. The population

Periodically, a sample of 30 tubes will be selected in standard deviation is believed to be 0.2 oz.

order to check the filling process.

Perform a hypothesis test, at the .03 level of

Quality assurance procedures call for the

significance, to help determine whether the filling

continuation of the filling process if the sample

process should continue operating or be stopped and

results are consistent with the assumption that the

mean filling weight for the population of toothpaste corrected.

tubes is 6 oz.; otherwise the process will be adjusted.

Two-Tailed Tests About a Population Mean: Two-Tailed Tests About a Population Mean:

s Known s Known

n p –Value and Critical Value Approaches n p –Value Approach

H a: µ ¹ 6

For z = 2.74, cumulative probability = .9969

2. Specify the level of significance. a = .03 p–value = 2(1 - .9969) = .0062

3. Compute the value of the test statistic. 5. Determine whether to reject H0.

z= = = 2.74

s / n .2 / 30 There is sufficient statistical evidence to

infer that the alternative hypothesis is true

(i.e. the mean filling weight is not 6 ounces).

7

Two-Tailed Tests About a Population Mean: Two-Tailed Tests About a Population Mean:

s Known s Known

n p-Value Approach n Critical Value Approach

p -value p -value

= .0031 = .0031 For a/2 = .03/2 = .015, z.015 = 2.17

Reject H0 if z < -2.17 or z > 2.17

a/2 = a/2 =

5. Determine whether to reject H0.

.015 .015

Because 2.74 > 2.17, we reject H0.

z There is sufficient statistical evidence to

z = -2.74 0 z = 2.74 infer that the alternative hypothesis is true

-za/2 = -2.17 za/2 = 2.17 (i.e. the mean filling weight is not 6 ounces).

s Known Two-Tailed Tests About a Population Mean

n Critical Value Approach n Select a simple random sample from the population

and use the value of the sample mean x to develop

Sampling the confidence interval for the population mean µ.

distribution (Confidence intervals are covered in Chapter 8.)

x - µ0

of z = n If the confidence interval contains the hypothesized

s/ n

value µ0, do not reject H0. Otherwise, reject H0.

Reject H 0 Do Not Reject H 0 Reject H 0 (Actually, H0 should be rejected if µ0 happens to be

equal to one of the end points of the confidence

a/2 = .015 a/2 = .015

interval.)

z

-2.17 0 2.17

Two-Tailed Tests About a Population Mean

In recent years, several companies have been formed to compete with

The 97% confidence interval for µ is Reliance JIO in unlimited calls data service. All advertise that their post-

s paid rates are lower than Reliance JIO’s, and as a result their bills will be

x ± za / 2 = 6.1 ± 2.17(.2 30) = 6.1 ± .07924 lower. Reliance JIO has responded by arguing that there will be no

n difference in billing for the average consumer and Jio’s data speed is

or 6.02076 to 6.17924 faster. Suppose that a data analyst working for Reliance JIO determines

that the mean and standard deviation of monthly post-paid bills for all

Because the hypothesized value for the its residential customers are $17.09 and $3.87, respectively. He then

population mean, µ0 = 6, is not in this interval, takes a random sample of 100 customers and recalculates their last

the hypothesis-testing conclusion is that the month’s bill using the rates quoted by a leading competitor. Assuming

that the standard deviation of this population is the same as for Reliance

null hypothesis, H 0: µ = 6, can be rejected. JIO, can we conclude at the 5% significance level that there is a

difference between the average Reliance JIO bill and that of the leading

competitor?

Data

8

Tests About a Population Mean: Tests About a Population Mean:

s Unknown s Unknown

■ Test Statistic ■ Rejection Rule: p -Value Approach

Reject H0 if p –value < a

x - µ0

t=

s/ n ■ Rejection Rule: Critical Value Approach

H0: µ > µ0 Reject H0 if t < -ta

This test statistic has a t distribution

with n - 1 degrees of freedom. H0: µ < µ0 Reject H0 if t > ta

n The format of the t distribution table provided in most ■ One-Tailed Test About a Population Mean: s Unknown

statistics textbooks does not have sufficient detail A State Highway Patrol periodically samples

to determine the exact p-value for a hypothesis test. vehicle speeds at various locations on a particular

n However, we can still use the t distribution table to roadway. The sample of vehicle speeds is used to

identify a range for the p-value. test the hypothesis H0: µ < 65.

n An advantage of computer software packages is that The locations where H0 is rejected are deemed the

the computer output will provide the p-value for the best locations for radar traps. At Location F, a

t distribution. sample of 64 vehicles shows a mean speed of 66.2

mph with a standard deviation of 4.2 mph. Use a

= .05 to test the hypothesis.

One-Tailed Test About a Population Mean: One-Tailed Test About a Population Mean:

s Unknown s Unknown

n p –Value and Critical Value Approaches n p –Value Approach

H a: µ > 65

For t = 2.286, the p–value must be less than .025

2. Specify the level of significance. a = .05 (for t = 1.998) and greater than .01 (for t = 2.387).

.01 < p–value < .025

3. Compute the value of the test statistic.

5. Determine whether to reject H0.

x - µ0 66.2 - 65

t= = = 2.286 Because p–value < a = .05, we reject H0.

s / n 4.2 / 64

We are at least 95% confident that the mean speed

of vehicles at Location F is greater than 65 mph.

9

One-Tailed Test About a Population Mean: One-Tailed Test About a Population Mean:

s Unknown s Unknown

n Critical Value Approach

Reject H 0 if t > 1.669 Reject H0

a = .05

CH012.qxd 11/22/10 8:14 PM Page 401

Because 2.286 > 1.669, we reject H 0.

We are at least 95% confident that the mean speed t

of vehicles at Location F is greater than 65 mph. 0 ta =

Location F is a good candidate for a radar trap. 1.669

INFERENCE ABOUT A POPULATION

401

manufacture aluminum cans from recycled cans than from bauxite. Newspapers are an

exception. It can be profitable to recycle newspaper. A major expense is the collection

from homes. In recent years, many companies have gone into the business of collecting

used newspapers from households and recycling them. A financial analyst for one such

company has recently computed that the firm would make a profit if the mean weekly

newspaper collection from each household exceeded 2.0 pounds. In a study to deter-

mine the feasibility of a recycling plant, a random sample of 148 households was drawn

from a large community, and the weekly weight of newspapers discarded for recycling

Example for each household was recorded and listedData next. Do these data provide sufficient evi-

In the near future, nations will likely have to do more to save the dence to allow the analyst to conclude that a recycling plant would be profitable?

environment. Possible actions include reducing energy use and recycling.

Currently, most products manufactured from recycled material are Weights of Discarded Newspapers

considerably more expensive than those manufactured from material 2.5 0.7 3.4 1.8 1.9 2.0 1.3 1.2 2.2 0.9 2.7 2.9 1.5 1.5 2.2

found in the earth. For example, it is approximately three times as 3.2 0.7 2.3 3.1 1.3 4.2 3.4 1.5 2.1 1.0 2.4 1.8 0.9 1.3 2.6

expensive to produce glass bottles from recycled glass than from silica 3.6 0.8 3.0 2.8 3.6 3.1 2.4 3.2 4.4 4.1 1.5 1.9 3.2 1.9 1.6

sand, soda ash, and limestone, all plentiful materials mined in numerous 3.0 3.7 1.7 3.1 2.4 3.0 1.5 3.1 2.4 2.1 2.1 2.3 0.7 0.9 2.7

countries. It is more expensive to manufacture aluminium cans from 1.2 2.2 1.3 3.0 3.0 2.2 1.5 2.7 0.9 2.5 3.2 3.7 1.9 2.0 3.7

recycled cans than from bauxite. Newspapers are an exception. It can be 2.3 0.6 0.0 1.0 1.4 0.9 2.6 2.1 3.4 0.5 4.1 2.2 3.4 3.3 0.0

profitable to recycle newspaper. A major expense is the collection from 2.2 4.2 1.1 2.3 3.1 1.7 2.8 2.5 1.8 1.7 0.6 3.6 1.4 2.2 2.2

homes. In recent years, many companies have gone into the business of 1.3 1.7 3.0 0.8 1.6 1.8 1.4 3.0 1.9 2.7 0.8 3.3 2.5 1.5 2.2

collecting used newspapers from households and recycling them. A 2.6 3.2 1.0 3.2 1.6 3.4 1.7 2.3 2.6 1.4 3.3 1.3 2.4 2.0

1.3 1.8 3.3 2.2 1.4 3.2 4.3 0.0 2.0 1.8 0.0 1.7 2.6 3.1

financial analyst for one such company has recently computed that the firm

would make a profit if the mean weekly newspaper collection from each

SOLUTION

household exceeded 2.0 pounds. In a study to determine the feasibility of a

recycling plant, a random sample of 148 households was drawn from a

large community, and the weekly weight of newspapers discarded for

recycling for each household was recorded and listed next. Do these data IDENTIFY

provide sufficient evidence to allow the analyst to conclude that a recycling The problem objective is to describe the population of the amounts of newspaper dis-

plant would be profitable? carded by each household in the population. The data are interval, indicating that the

parameter to be tested is the population mean. Because the financial analyst needs to

determine whether the mean is greater than 2.0 pounds, the alternative hypothesis is

H1: m 7 2.0

As usual, the null hypothesis states that the mean is equal to the value listed in the alter-

native hypothesis:

H0: m = 2.0

A Summary of Forms for Null and Alternative The test statistic isTests About a Population Proportion

Hypotheses About a Population Proportion x - m

t = n = n- 1

■The equality part of the hypotheses always appears ■ s>Test

2n Statistic

in the null hypothesis. p - p0

n In general, a hypothesis test about the value of a z=

COMPUTE sp

population proportion p must take one of the

following three forms (where p0 is the hypothesized MANUALLY where:

value of the population proportion). The manager believes that the cost of a Type I error (concluding that the mean is

greater than 2 when it isn’t) is quite high. p0 (1 - p0 )he sets the significance level at

s p =Consequently,

1%. The rejection region is n

H 0: p > p0 H 0: p < p0 H 0: p = p0

t 7 ta,n = t.01,148 L t.01,150 = 2.351

H a: p < p0 H a: p > p0 H a: p ≠ p0

assuming np > 5 and n(1 – p) > 5

One-tailed One-tailed Two-tailed

Copyright 2011 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).

(lower tail) (upper tail) Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.

10

Tests About a Population Proportion Two-Tailed Test About a

Population Proportion

■ Rejection Rule: p –Value Approach ■ Example: National Safety Council (NSC)

Reject H 0 if p –value < a For a Christmas and New Year’s week, the

National Safety Council estimated that 500 people

■ Rejection Rule: Critical Value Approach

would be killed and 25,000 injured on the nation’s

H 0: p < p0 Reject H 0 if z > za roads. The NSC claimed that 50% of the accidents

H 0: p > p0 Reject H 0 if z < -za would be caused by drunk driving.

A sample of 120 accidents showed that 67 were

H 0: p = p0 Reject H 0 if z < -za/2 or z > za/2

caused by drunk driving. Use these data to test the

NSC’s claim with a = .05.

Population Proportion Population Proportion

n p –Value and Critical Value Approaches n p-Value Approach

H a : p ¹ .5 For z = 1.28, cumulative probability = .8997

2. Specify the level of significance. a = .05 p–value = 2(1 - .8997) = .2006

3. Compute the value of the test statistic. 5. Determine whether to reject H0.

Because p–value = .2006 > a = .05, we cannot reject H0.

p0 (1 - p0 ) .5(1 - .5)

a common

sp = = = .045644

n 120

error is using

p in this p - p0 (67 /120) - .5

formula z= = = 1.28

sp .045644

Population Proportion

n Critical Value Approach

Reject H 0 if z < -1.96 or z > 1.96

11