Sie sind auf Seite 1von 12

3.

Source of
Variation

SS
220.7
0
2,252.2
0
2,472.9
0

Between Groups
Within Groups
Total
At

df
2

110.35

21

107.25

F crit at 1%

1.03

5.78

23

0.01,using the table we find F ,(df

F(2,21)=1.03<5.78

MS

=F 0.01,(2,21)=5.78 . Since

,df 2 )

, we do not reject

H 0 . At the 1% significance

level, we cannot conclude that the population means are not all equal.
4.

Source of
Variation
Between Groups

SS

df

MS

11.3

3.78

59.1

56

1.06

70.4

59

F crit at 5%

3.58

2.77

4
Within Groups
3
Total
7
At

=0.05, F ,(df

reject

=F 0.05,(3,56) =2.77 . Since

F(3,56)=3.58> 2.77,

,df 2 )

we

H 0 . At the 5% significance level, we can conclude that the

population means are not all equal.


10.
a.

H 0 : Snorkeling =Sailing = NBoarding/Windsurfing =Bowling =Onroad triathlon =Off road triathlon


H A : Not all population meanincomes are equal .

Source of
Variation
Between

SS
4,895

df

MS

979.03

37.85

pvalue
0.000

F crit at 5%
2.62

Groups
Within Groups
Total

.15
62
0.80
5,515
.95

24

25.87

29

Note: p-value is obtained from Excel using F.DIST.RT()

b. At =0.05,
c.

F ,(df

=F 0.05,(5,24 )=2.62 .

, df 2)

Since F(5,24) =37.85>2.62,

we reject

So , wereject H 0 if F(5,24 ) >2.62

H 0 . At the 5% significance

level, we can conclude that average incomes differ depending on the


recreational sport.

11.
ANOVA: Single Factor
SUMMARY
Groups
Section 1
Section 2
Section 3

Count
10
10
10

ANOVA
Source of
Variation
Between
Groups

SS
442.8
7
2858.
10
3300.
97

Within Groups
Total
At = 0.05,

F ,(df

Sum
728
787
694

df
2
27

Avera
ge
72.8
78.7
69.4

Varianc
e
142.84
104.01
70.71

MS
221.4
3
105.8
6

Pvalue

F crit

2.09

0.143

3.35

29

=F 0.05,(2,27) =3.35

, df 2)

So ,the decision rule is reject H 0 if F(2,27) >3.35 .


Since F(2,27)=2.09<3.35, we

do not reject

H 0 . At the 5% significance

level, we cannot conclude that average final exam scores are not
equal among the instructors three sections.
13.
a. Tukeys HSD approach
Population Mean
Differences

12
13
1 4

Confidence Interval: (
( 149154 ) 4.70
( 149143 ) 4.70
( 149139 ) 4.70

x ix j ) q ,(c ,n c) MSE
n
T

51.30
=[ 15.76,5.76 ]
10
51.30
=[4.76, 16.76]
10
51.30
=[0.76, 20.76]
10

23
2 4

( 154143 ) 4.70
( 154139 ) 4.70

51.30
=[ 0.24,21.76 ] *
10
51.30
=[ 4.24, 25.76 ] *
10

3 4

( 143139 ) 4.70

At =0.01,

51.30
=[6.76, 14.76]
10

c=4,n T c=404=36, so q ,(c ,n c)=q0.01,(4,36)=4.75


T

Intervals not containing the value zero are indicated by the asterisk *.
b. All the confidence intervals contain the value zero, with the exception
of

23

and

2 4 . Intervals not containing the zero value indicate

that the corresponding means are different at the 1% significance


level. Specifically, the mean of population 2 significantly differs from
that of population 3; it also differs from that of population 4.
14.
a.

H 0 : 1=2=3 ; H A : Not all population means are equal.


Since the p-value 0.0006 <

0.05= , we reject

H 0 . At the 5%

significance level, we conclude that the population means are not all
equal.
b.
Tukeys HSD approach
Population Mean
Differences

12
13
23

At =0.05,

x ix j ) q ,(c ,n c) MSE
n

Confidence Interval: (

( 0. 571 .3 8 ) 3.67

0.36
=[ 1.71,0.09 ]
6

( 0. 572.33 ) 3.67

0.36
=[2.66,0.86 ] *
6

( 1.3 82.33 ) 3.67

0.36
=[1.85,0.05 ] *
6

c=3, nT c=183=15, so q ,(c ,n c) =q0.05,(3,15)=3.67


T

Intervals not containing the value zero are indicated by the asterisk *;
they represent population mean differences that differ at the 5%
significance level.

c. At the 5% significance level, we can conclude that

and that

2 differs from

differs

; we cannot conclude that

from

2 significantly differ from one another.

and
20.

Anova: Single Factor


SUMMARY
Groups
August 31
November 30
February 28
May 31

Count
10
10
10
10

Sum
37576
33669
34301
38624

Average
3757.6
3366.9
3430.1
3862.4

Variance
995333.6
802284.1
949813.9
913229.6

df

MS

Within Groups

SS
176820
9
329459
51

Total

347141
60

ANOVA
Source of Variation
Between Groups

589402.97

36

915165.29

0.644

P-value

F crit

0.5918

2.866

39

H 0 : 1=2=3= 4 ; H A : Not all population means are equal .


At

=0.05, F ,(df

=F 0.05,(3,36) =2.866 . Since

,df 2 )

F(3,36)=0.644 <2.866 ,

we donot reject

H 0 . At the 5% significance level, we cannot conclude that the average revenue


for the four quarters are not equal .
It is not necessary to construct confidence intervals using Tukeys approach since
we were unable to reject the null hypothesis in part a. In other words, no
significant differences between the means will be found.
23.
a.
Source of
Variation

SS

df

MS

F crit at 5%

Rows
Columns
Error
Total

40.80
201.60
69.30
311.70

4
2
8
14

10.20
100.80
8.66

1.18
11.64

3.84
4.46

SSB=SST ( SSA+ SSE )=311.7( 201.6+69.3 )=40.8


b. At

0.05, F ,(df

not reject

,df 2)

=F0.05,(4,8) =3.84

F(4,8) =1.18<3.84

, thus, we do

H 0 . At the 5% significance level, we cannot conclude that

the row means significantly differ.


c. At

0.05, F ,(df

,df 2)

=F0.05,(2,8)=4.46

F(2,8)=11.64 > 4.46

, we reject

H 0 . At the 5% significance level, we can conclude that the column


means significantly differ.
25.
1.
a.
Source of
Variation
Rows
Columns
Error
Total

SS

df

MS

p-value

F crit at 5%

25.17
142.25
19.5
186.92

2
3
6
11

12.59
47.42
3.25

3.87
14.59

0.0832
0.0037

5.143
4.757

Note: p-value is obtained from Excel using F.DIST.RT()

b. Since the p-value=

0.0037<0.05= , we reject

H 0 . At the 5%

significance level, we can conclude that the column means


significantly differ.

Since the p-value=

0.0832>0.05= , we do not reject

H 0 . At the 5% significance

level, we cannot conclude that the row means significantly differ.


30.
a.
Anova: Two-Factor Without Replication
SUMMARY
Count
Su Averag

Varianc

Cape Cod

Colonial

Ranch

ANOVA
Source of
Variation
Rows
Columns
Error
Total

m
127
0
162
0
117
0
134
5
136
5
135
0

SS
37222.
22
72.22
444.44
37738.
89

423.33

58.33

540.00

100.00

390.00

100.00

448.33
455.00
450.00

df

MS
18611.
11
36.11
111.11

2
2
4

5308.3
3
6825.0
0
6700.0
0

P-value

F crit

167.50
0.32

0.0001
0.7400

6.944
6.944

b. Since the p-value = 0.74>

0.05= ,

we do not reject

H 0 . At the

5% significance level, we cannot conclude that the average values


differ by appraiser, suggesting that the appraisers are consistent with
their valuations.
c. Since the p-value = 0.0001<

=0.05 , we reject

H 0 . At the 5%

significance level, average values significantly differ by house type.


d. Tukeys HSD approach
Population Mean
Differences

12

( 423.3540 ) 4.34

x ix j ) q ,(c ,n c) MSE
n

Confidence Interval: (

111.11
= [143.08,90.26 ]
3

111.11
=[6.92,59.74 ] *
3

13

( 423.3390 ) 4.34

23

123.59,176 .41 }
( 540390 ) 4.34 111.11 = *
3

At =0.01,

c=3, nT c=93=6, so q ,(c, n c)=q 0.05, (3,6 )=4.34


T

Intervals not containing the value zero are indicated by the asterisk *.
At the 5% significance level, the population means for house type all
differ.

31.
a.
ANOVA: Two-Factor With Replication
Source of
Variation
Sample (Rows)
Columns
Interaction
Within (Error)
Total

SS

df
400
1,008
30
120
1,558

MS
4
2
8
75
89

F
62.50
315
2.34

100
504
3.75
1.60

b. Since the p-value (interaction) = 0.026> 0.01 =

pvalue
0.000
0.000
0.026

F crit at
1%
3.58
4.90
2.76

we do not reject

H 0 . At the 1% significance level, we cannot conclude that there is


interaction between factors A and B.
c. Since the p-value (columns)

< 0.01 =

we reject

H 0 . At

the 1% significance level, average values differ for factor A.


d. Since the p-value (rows)

0 < 0.01 =

we reject

H 0 . At the

1% significance level, average values differ for factor B.


33.
a.
ANOVA: Two-Factor With Replication
Source of
Variation

F crit at 5%
SS

df

MS

p-value

Sample (Rows)
Columns
Interaction
Within (Error)
Total

752.78
12,012.5
0

58.33
683.33
13,506.9
4

2
12

376.39
12,012.
50
29.17
56.94

6.61

0.0116

210.95

0.0000

0.51

0.6117

3.885
4.747
3.885

17

b. Since the p-value (interaction)= 0.61 > 0.05= ,

we do not reject

H 0 . At the 5% significance level, we cannot conclude that there is


interaction between factors A and B.
c. Since the p-value (columns)

0<0.05= ,

we reject

H 0 . At the 5%

significance level, the column means differ.


d. Since the p-value (rows) = 0.0116 <

0.05= ,

H 0 . At the

we reject

5% significance level, the row means differ.


35.
a. Since the p-value (interaction)= 0.078 0.05= , we do not reject

H 0 . At the 5% significance level, we cannot conclude that there is


interaction between the detergent brand name and the temperature
of the water.
b. Since the interaction between the between the detergent brand name
and the temperature of the water is not significant at the 5%
significance level, we can determine whether the means differ by
detergent brand name (factor A) and water temperature ( factor B),
respectively.

11

Since the p-value (columns) = 1.45 10


reject

<

0.05= , we

H 0 . At the 5% significance level, differences exist in

the average brightness of fabrics depending on the detergent


brand name.

10

Since the p-value (columns) = 8.92 10


reject

<

0.05= , we

H 0 . At the 5% significance level, differences exist in

the average brightness of fabrics depending on the


temperature of the water.

36.
a. Relevant Excel Output

ANOVA
Source of Variation
Sample
Columns
Interaction
Within
Total

SS
20.05
6
14.11
1
0.111
4.000
38.27
8

df
1

MS
20.05
6

2
2
12

7.056
0.056
0.333

F crit

0.000

4.747

0.000
0.848

3.885
3.885

0.05= , we

do not

F
60.16
7
21.16
7
0.167

17

b. Since the p-value (interaction) = 0.848 >


reject

Pvalue

H 0 . At the 5% significance level, we cannot conclude

that there is interaction between the brand name and whether


the car is garaged.
c. Since the p-value (columns)

0 <

0.05= , we reject

H0 .

At the 5% significance level, we can conclude that average


battery lives significantly differ by brand name.
d. Since the p-value (rows)

0 <

0.05= , we reject

H 0 . At

the 5% significance level, we can conclude that average battery


lives differ depending on whether a car is garaged or not.

38.
a. Relevant Excel Output
ANOVA
Source of
Variation
Sample
Columns
Interaction

SS
383.6
3
2591.
19
622.5
9

df
2
2
4

MS
191.81
5
1295.5
93
155.64
8

Pvalue

F crit

1.844
12.45
3

0.187

3.555

0.000

3.555

1.496

0.245

2.928

1872.
67
5470.
07

Within
Total

18

104.03
7

26

b. Since the p-value (interaction)= 0.245>


reject

0.05= , we

do not

H 0 . At the 5% significance level, we cannot conclude

that there is interaction between industry and work experience.


c. Since the p-value (columns)

<

0.05= , we reject

H0 .

At the 5% significance level, we can conclude that job


satisfaction differs by industry.
d. Since the p-value (rows) = 0.187 >

0.05= , we do not reject

H 0 . At the 5% significance level, we cannot conclude job


satisfaction differs depending on work experience.

Das könnte Ihnen auch gefallen