mat 157 ch. 8

33
Chapter 8 Hypothesis Test

Upload: arisil23

Post on 07-Apr-2018

221 views

Category:

Documents


2 download

TRANSCRIPT

Page 1: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 1/33

Chapter 8

Hypothesis Test

Page 2: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 2/33

Steps to a Hypothesis Test

1. Hypotheses

 ± Null Hypothesis (Ho)

 ± Alternative Hypothesis (Ha)2. Alpha

3. Distribution (aka model)

4. Test Statistics and P-value5. Decision

6. Conclusion

Page 3: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 3/33

Steps to a Hypothesis Test

� Can remember the steps by the sentence:

³Happy AuntsMake The Darndest

Cookies´

Page 4: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 4/33

Example 1± Hypothesis Testing

� An attorney claims that more than 25% of 

all lawyers advertise. A sample of 200

lawyers in a certain city showed that 63 had

used some form of advertising. At = 0.05,

is there enough evidence to support the

attorney¶s claim?

Page 5: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 5/33

Hypotheses (Sets up the two

sides of the test)1. Build the Alternative Hypothesis (Ha) first.

 ± based on the claim you are testing (you get

this from the words in the problem)

� Three choices

 ± Ha: parameter � hypothesized value

 ± Ha: parameter < hypothesized value

 ± Ha: parameter > hypothesized value

2. Build Null Hypothesis (Ho) next.

 ± opposite of the Ha (i.e. = , � , � )

Page 6: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 6/33

Example 1± Constructing

Hypotheses� We need to know what parameter we are

testing and which of the three choices for 

alternative hypothesis we are going to use.

 ± ³An attorney claims that more than 25% of all

lawyers advertise´ tells us that this is a test for 

 proportions so our parameter is p. ± ³claims that more than 25%´ tells us that

Ha: p > .25 and thereforeHo: p � .25

Page 7: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 7/33

Alpha� Alpha = = significance level

 ± How much proof we are requiring in order to

reject the null hypothesis.

 ± The complement of the confidence level that

we learned in the last chapter 

 ± Usually given to you in the problem, if not, youcan choose.

� Most popular alphas: 0.05, 0.01, and 0.10

Page 8: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 8/33

Example 1 ± Alpha

� ³At = 0.05´ is given to us in the problem

so we just copy = 0.05

Page 9: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 9/33

Model� The model is the distribution used for the

 parameter that you are testing. These are

 just the same as we used in the confidence

intervals.

 ± p and (n � 30) use the normal distribution

 ± (n < 30) uses the t-distribution ± uses the chi-squared distribution2

Page 10: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 10/33

Example 1 - Model

� The model used for a proportion is the

normal.

Page 11: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 11/33

TestS

tatistic� You will have a different test statistic for 

each of the four different parameters that we

have learned about.

 ± p :

 ± (n � 30) :

n

q p

 p p z

oo

o!

Ö

n

 x z o

 Q!

Page 12: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 12/33

TestS

tatistic� You will have a different test statistic for 

each of the four different parameters that we

have learned about.

 ± (n < 30) :

 ± :

n

 s

 xt  o

ndf  

 Q!

! )1(

2W  2

2

1)n(df 2 )1(

W  G 

 sn !!

Page 13: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 13/33

 p-value� This is the evidence (probability) that you

will get off of your chart and then compare

against your criteria (alpha).

� You will need to find the appropriate

 probability that goes with your Ha.

 ± > and < Ha¶s are called one-tailed tests.

 ± � Ha¶s are called two-tailed tests.

� For z and  2 you have to take the > probability X2

Page 14: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 14/33

Example 1 ± Test Statistic and

 p-value� The formula for a test statistic for 

 proportions is:

� So, from our problem we need a proportion

from a sample (p-hat), the proportion from

our hypothesis (po), and a sample size (n).

n

q p

 p p z

oo

o

!

Ö

Page 15: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 15/33

Example 1 ± Test Statistic and

 p-value� ³A sample of 200 lawyers in a certain city

showed that 63 had used some form of 

advertising´ tells us that

 ± p-hat = 63/200 or 0.315

� From our hypothesis we know

 ± po = 0.25 (which means that qo = 0.75)

� ³sample of 200´ tells us that

 ± n = 200

Page 16: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 16/33

Example 1 ± Test Statistic and

 p-value� So our test statistic and p-value are

value.- pour isnumber this

 0.01702.12)(z

 isaour insignthe because

 2.12)(zneedevalue- por 

12.2

200

)75.0)(25.0(

25.0315.0

!"

"

"

!

! z

Page 17: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 17/33

Decision ± (always about Ho)� We have two choices for decision

 ± Reject Ho

 ± Do Not Reject Ho

� If our evidence (p-value) is less than we

REJECT Ho.

� If our evidence (p-value) is greater than

we DO NOT REJECT Ho.

Page 18: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 18/33

Example 1 - Decision

� Our p-value is 0.0170 and our alpha is 0.05

 ± So, since our p-value is less than our alpha our 

decision is: REJECT Ho.

Page 19: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 19/33

Page 20: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 20/33

Page 21: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 21/33

Example 1 -

Summary

1. Ho: p � 0.25

Ha: p > 0.25

2. = 0.053. Model: Normal

4. z = 2.12 and p-value = 0.0170

5. Reject Ho

6. There is enough evidence to suggest that

 p>0.25.

Page 22: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 22/33

Example 2 ± Hypothesis Testing

A researcher reports that the average salary of 

assistant professors is more than $42,000. A

sample of 30 assistant professors has amean of $43,260. At = 0.05, test the claim

that assistant professors earn more than

$42,000 a year. The standard deviation of the population is $5230.

Page 23: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 23/33

Example 2 (cont.)

� Hypotheses

 ± Ho: � $42,000

 ± Ha: > $42,000 (given claim is ³more than´)

� Alpha

 ± = 0.05 (given)

� Model

 ± Normal (n � 30 and it¶s a mean)

Page 24: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 24/33

Example 2 (cont.)

� Test statistic and p-value:

0934.09066.011.32)(z:value-

32.1

305230$

000,42$260,43$

!!"

!

!

!

 z

n

 x z W 

 Q

Page 25: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 25/33

Example 2 (cont.)

� Decision

 ± 0.0934 > 0.05 (p-value > alpha)

 ± DO NOT R EJECT Ho

� Conclusion

 ± We do not have evidence to suggest that

> $42,000.

Page 26: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 26/33

Example 3 ± Hypothesis Testing

A physician claims that joggers¶ maximal

volume oxygen uptake is greater than the

average of all adults. A sample of 15 joggers has a mean of 40.6 milliliters per 

kilogram (ml/kg) and a standard deviation

of 6 ml/kg. If the average of all adults is36.7 ml/kg, is there enough evidence to

support the physicians claim at = 0.05?

Page 27: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 27/33

Example 3 (cont.)

� Hypotheses

 ± Ho: � 36.7

 ± Ha: > 36.7

� Alpha

 ± = 0.05 (given)

� Model

 ± t(14)

Page 28: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 28/33

Example 3 (cont.)

� Test statistic and p-value:

0.025.and0.01 between2.517)P(t

on)distributi-tfor therangea be(willvalue-P

517.2

15

6

7.366.40

(14) !"

!

!

!

n s

 xt 

Q

Page 29: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 29/33

Example 3 (cont.)

� Decision

 ± (0.01,0.025) < 0.05 (p-value < alpha)

 ± R EJECT Ho

� Conclusion

 ± There is evidence to suggest that > 36.7.

Page 30: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 30/33

Page 31: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 31/33

Ex

ample 4 (cont.)� Hypotheses

 ± Ho: = 16.8

 ± Ha: � 16.8

� Alpha

 ± = 0.05 (given)

� Model

 ±  2(23)

Page 32: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 32/33

Ex

ample 4 (cont.)� Test statistic and p-value:

0.10)(0.05,0.05)(0.025,*212.733))(P(*2

on)distributi-for therangea be(willvalue-P

733.12

)8.16(

)5.12)(124(

)1(

(23)2

2

2

22

2

22

!!{

!

!

!

 G 

 G 

 G 

 G   sn

Page 33: MAT 157 Ch. 8

8/6/2019 MAT 157 Ch. 8

http://slidepdf.com/reader/full/mat-157-ch-8 33/33

Ex

ample 4 (cont.)� Decision

 ± (0.05,0.10) > 0.05 (p-value > alpha)

 ± DO NOT R EJECT Ho

� Conclusion

 ± There is not enough evidence to suggest that

� 16.8.