Appendix 15

A15.1 Chi-squared goodness-of-fit test

, ,

At least one is not equal to its specified value.

C D E
1 Actual Expected
2 183 175
3 63 70
4 55 52.5
5 29 35
6 20 17.5
7 p-value= 0.6321
p-value = .6321. There is not enough evidence to conclude that applicants to WLU's MBA program are different in

terms of their undergraduate degrees from the population of MBA applicants?

A15.2 t-test of

=0

<0

A B C
1 t-Test: Paired Two Sample for Means
2
3 First Sat Second SAT
4 Mean 1175 1190
5 Variance 28422 35392
6 Observations 40 40
7 Pearson Correlation 0.91
8 Hypothesized Mean Difference 0
9 df 39
10 t Stat -1.20
11 P(T<=t) one-tail 0.1182
12 t Critical one-tail 1.6849
13 P(T<=t) two-tail 0.2365
14 t Critical two-tail 2.0227

© 2015 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a
license distributed with a certain product or service or otherwise on a password-protected website for classroom use.
t = –1.20, p-value = .1182. There is not enough evidence to indicate that repeating the SAT produces higher exam
scores.
A15.3 Time to solve the 48 problems: Equal-variances t-test of

A B C
1 t-Test: Two-Sample Assuming Equal Variances
2
3 Diet Not
4 Mean 581.95 551.5
5 Variance 2716.6 2221.5
6 Observations 20 20
7 Pooled Variance 2469.1
8 Hypothesized Mean Difference 0
9 df 38
10 t Stat 1.94
11 P(T<=t) one-tail 0.0300
12 t Critical one-tail 1.6860
13 P(T<=t) two-tail 0.0601
14 t Critical two-tail 2.0244
t = 1.94, p-value = .0300. There is enough evidence to conclude that dieters take longer to solve the 48 problems
than do nondieters.

Successfully repeat string of five letters: z-test of (case 1)

=0
<0

© 2015 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a
license distributed with a certain product or service or otherwise on a password-protected website for classroom use.
 A B C D
1 z-Test: Two Proportions
2
3 Diet Not
4 Sample Proportions 0.50 0.80
5 Observations 20 20
6 Hypothesized Difference 0
7 z Stat -1.99
8 P(Z<=z) one tail 0.0234
9 z Critical one-tail 1.6449
10 P(Z<=z) two-tail 0.0468
11 z Critical two-tail 1.96
z = –1.99, p-value = .0234. There is enough evidence to conclude that dieters are less successful at repeating string
of five letters.

Successfully repeat string of five words: z-test of (case 1)

=0
>0

A B C D
1 z-Test: Two Proportions
2
3 Diet Not
4 Sample Proportions 0.35 0.60
5 Observations 20 20
6 Hypothesized Difference 0
7 z Stat -1.58
8 P(Z<=z) one tail 0.0567
9 z Critical one-tail 1.6449
10 P(Z<=z) two-tail 0.1134
11 z Critical two-tail 1.96
z = –1.58, p-value = .0567. There is not enough evidence to conclude that dieters are less successful at repeating
string of five words.

A15.4 a t-estimator of

© 2015 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a
license distributed with a certain product or service or otherwise on a password-protected website for classroom use.
 A B C D
1 t-Estimate: Mean
2
3 Overdue
4 Mean 7.09
5 Standard Deviation 6.97
6 LCL 6.40
7 UCL 7.77
LCL = 6.40, UCL = 7.77
b LCL = 50,000($.25)(6.40) = $80,000
UCL = 50,000($.25)(7.77) = $97,125
It does appear that not all fines are collected

A15.5 One-way analysis of variance

At least two means differ

A B C D E F G
10 ANOVA
11 Source of Variation SS df MS F P-value F crit
12 Between Groups 57512 2 28756 3.23 0.0468 3.16
13 Within Groups 506984 57 8894
14
15 Total 564496 59

F = 3.23; p-value = .0468. There is enough evidence to conclude that differences in sales exist between the three
advertising strategies.

A15.6 Chi-squared test of a contingency table

The two variables (income category and mutual fund ownership) are independent

The two variables are dependent

© 2015 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a
license distributed with a certain product or service or otherwise on a password-protected website for classroom use.
 A B C D E
1 Contingency Table
2
3 Income category
4 Mutual fund 1 2 TOTAL
5 1 71 13 84
6 2 59 28 87
7 3 86 55 141
8 4 87 157 244
9 5 32 145 177
10 6 58 205 263
11 TOTAL 393 603 996
12
13
14 chi-squared Stat 196.77
15 df 5
16 p-value 0
17 chi-squared Critical 11.0705

= 196.77;p-value = 0. There is overwhelming evidence to infer that household income and ownership of mutual
funds are related

A15.7 Two-factor analysis of variance

A B C D E F G
23 ANOVA
24 Source of Variation SS df MS F P-value F crit
25 Sample 13172 1 13172 1.42 0.2387 4.02
26 Columns 98839 2 49419 5.33 0.0077 3.17
27 Interaction 1610 2 805 0.09 0.9171 3.17
28 Within 501137 54 9280
29
30 Total 614757 59

Interaction: F = .09; p-value = .9171. There is no evidence of interaction.

Advertising strategy (Columns): F = 5.33; p-value = .0077. There is sufficient evidence to conclude that advertising
strategies differ with respect to sales.
Media (Sample): F = 1.42; p-value = .2387. There is not enough evidence to conclude that differences in the
medium for advertising differ in terms of sales.

A15.8 z-test of (case 1) Code 3 results were omitted.

© 2015 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a
license distributed with a certain product or service or otherwise on a password-protected website for classroom use.
 A B C D
1 z-Test: Two Proportions
2
3 Folic acid Placebo
4 Sample Proportions 0.0101 0.0343
5 Observations 597 612
6 Hypothesized Difference 0
7 z Stat -2.85
8 P(Z<=z) one tail 0.0022
9 z Critical one-tail 1.6449
10 P(Z<=z) two-tail 0.0044
11 z Critical two-tail 1.9600
z = –2.85, p-value = .0022. There is overwhelming evidence to conclude that folic acid reduces the incidence of
spina bifida.

A15.9 Unequal-variances t-test of

=0

<0

A B C
1 t-Test: Two-Sample Assuming Unequal Variances
2
3 British American
4 Mean 238.0 252.0
5 Variance 149.9 220.2
6 Observations 263 279
7 Hypothesized Mean Difference 0
8 df 531
9 t Stat -12.00
10 P(T<=t) one-tail 7.64E-30
11 t Critical one-tail 1.6477
12 P(T<=t) two-tail 1.53E-29
13 t Critical two-tail 1.9644

z = –12.00, p-value = 0. There is enough evidence to infer that British golfers play golf in less time than do
American golfers.

© 2015 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a
license distributed with a certain product or service or otherwise on a password-protected website for classroom use.
A15.10 one-way analysis of variance

At least two means differ

A B C D E F G
10 ANOVA
11 Source of Variation SS df MS F P-value F crit
12 Between Groups 626046 2 313023 58.37 0.0000 3.03
13 Within Groups 1523047 284 5363
14
15 Total 2149093 286
F = 58.37, p-value = 0. There is enough evidence to conclude that there are differences between the three groups.

Multiple Comparisons
A B C D E
1 Multiple Comparisons
2
3 LSD Omega
4 Treatment Treatment Difference Alpha = 0.0167 Alpha = 0.05
5 Before 1976 After 1986 122.62 28.03 25.46
6 Canadian 78.04 24.67 25.46
7 After 1986 Canadian -44.58 25.75 25.46
All three groups differ from each other.

A15.11 Chi-squared test of a contingency table

The two variables (year and sport) are independent

The two variables are dependent

© 2015 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a
license distributed with a certain product or service or otherwise on a password-protected website for classroom use.
 A B C D E
1 Contingency Table
2
3 Sport
4 Year 1 2 TOTAL
5 1 116 122 238
6 2 119 92 211
7 3 29 58 87
8 4 52 39 91
9 5 48 34 82
10 6 16 33 49
11 7 26 29 55
12 8 24 21 45
13 9 70 72 142
14 TOTAL 500 500 1000
15
16
17 chi-squared Stat 23.8101
18 df 8
19 p-value 0.0025
20 chi-squared Critical 15.5073

= 23.8101, p-value = .0025. There is enough evidence to infer that North Americans changed their favorite sport
between 1985 and 1992.

A15.12 t-estimator of

A B C D
1 t-Estimate: Mean
2
3 Cars
4 Mean 165.79
5 Standard Deviation 51.59
6 LCL 157.17
7 UCL 174.41

Five minute interval: LCL = 157.17, UCL = 174.41

Twenty-four hour day (12 5-minute intervals, 24 hours per day):
LCL = 12 24 157.17 = 45,265
UCL = 12 24 174.41 = 50,230

A15.13 z-estimator of p

© 2015 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a
license distributed with a certain product or service or otherwise on a password-protected website for classroom use.
 A B
1 z-Estimate: Proportion
2 Exercise?
3 Sample Proportion 0.551
4 Observations 671
5 LCL 0.514
6 UCL 0.589
Total number of adults who exercise:
LCL = 205.9 million (.514) = 105.8 million
UCL = 205.9 million (.589) = 121.3 million

A15.14 Equal-variances t-test of

<0

A B C
1 t-Test: Two-Sample Assuming Equal Variances
2
3 Activity Usual
4 Mean 57.06 87.28
5 Variance 296.18 215.42
6 Observations 67 67
7 Pooled Variance 255.80
8 Hypothesized Mean Difference 0.00
9 df 132
10 t Stat -10.94
11 P(T<=t) one-tail 0.0000
12 t Critical one-tail 1.6565
13 P(T<=t) two-tail 0.0000
14 t Critical two-tail 1.9781
t = –10.94, p-value = 0. There is enough evidence to indicate to infer that graded activity is effective.

A15.15 Chi-squared test of a contingency table

The two variables (group and improvement) are independent

The two variables are dependent

© 2015 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a
license distributed with a certain product or service or otherwise on a password-protected website for classroom use.
 A B C D E
1 Contingency Table
2
3 Group
4 Improvement 1 2 TOTAL
5 1 42 8 50
6 2 32 18 50
7 3 13 37 50
8 TOTAL 87 63 150
9
10
11 chi-squared Stat 35.63
12 df 2
13 p-value 0
14 chi-squared Critical 5.9915

= 35.63, p-value = 0. There is sufficient evidence to infer there are differences between the three groups.

A15.16 z-estimator of p

A B C D E
1 z-Estimate of a Proportion
2
3 Sample proportion 0.774 Confidence Interval Estimate
4 Sample size 780 0.774 ± 0.0294
5 Confidence level 0.95 Lower confidence limit 0.7446
6 Upper confidence limit 0.8034

Total number of on-time departures:

LCL = 7,140,596(.7446) = 5,316,888
UCL = 7,140,596(.8034) = 5,736,755

A15.17 =0

≠0

F = 1.00, p-value .9902 Use equal variances t-test.

© 2015 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a
license distributed with a certain product or service or otherwise on a password-protected website for classroom use.
t = 3.10, p-value = .0019. There is enough evidence to infer that job tenure changed between 2006 and 2008.

A15.18 Chi-squared test of a contingency table

The two variables (party and support for capital punishment) are independent

The two variables are dependent

110.35, p-value = 0. There is enough evidence to conclude that Democrats, Republicans, and Independents
differ in their support for capital punishment.

© 2015 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a
license distributed with a certain product or service or otherwise on a password-protected website for classroom use.
A15.19 t-estimate of a mean

LCL = 23.49, UCL = 24.04

A15.20 Chi-squared test of a contingency table

The two variables (PARTYID and SEX) are independent

The two variables are dependent

41.27, p-value = 0. There is sufficient evidence to conclude that men and women differ in their political
affiliation.

A15.21 one-way analysis of variance

© 2015 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a
license distributed with a certain product or service or otherwise on a password-protected website for classroom use.
 = µ4

At least two means differ

F = 21.86, p-value = 6.31E-14 ≈ 0. There is enough evidence to conclude that there are differences between classes
with respect to the number of family earners.

A15.22 one-way analysis of variance

= µ4

At least two means differ

© 2015 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a
license distributed with a certain product or service or otherwise on a password-protected website for classroom use.
F = 14.45, p-value = 5.5E-07 ≈ 0. There is enough evidence to infer that inflation-adjusted incomes differ between
2002, 2004, and 2006.

A15.23 Chi-squared test of a contingency table

The two variables (RACE and WRKSLF) are independent

The two variables are dependent

13.87, p-value = .0010. There is sufficient evidence to conclude that differences exist between the races in
whether an individual is self-employed.
A15.24 Chi-squared test of a contingency table
The two variables (PARTYID3 and UNEMP) are independent

The two variables are dependent

© 2015 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a
license distributed with a certain product or service or otherwise on a password-protected website for classroom use.
 14.61, p-value = .0007. There is sufficient evidence to conclude that Americans who have been unemployed in
the last ten years have different party affiliations.

A15.25 H0: (µ1 - µ2) = 0

H1: (µ1 - µ2) ≠ 0
F = .866 p-value = .1339. Use equal-variances

t =.600, p-value = .5485. There is not enough evidence to infer that Liberals and Conservatives differ in their
intention to vote.

A15.26 Chi-squared test of a contingency table

The two variables (Year and EMPLOY) are independent

The two variables are dependent

© 2015 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a
license distributed with a certain product or service or otherwise on a password-protected website for classroom use.
 2077, p-value = 0. There is sufficient evidence to conclude that employment status changed between 2004 and
2008.

A15.27 H0: (µ1 - µ2) = 0

H1: (µ1 - µ2) > 0

F = 1.14, p-value = .1429. Use equal-variances t-test

t = .152, p-value = .8795. There is not sufficient evidence to conclude that American who consider themselves
strong Democrats or Republicans have more education than those who are not strong Democrats or Republicans.

© 2015 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a
license distributed with a certain product or service or otherwise on a password-protected website for classroom use.
Case A15.1 One-way analysis of variance

At least two means differ

Weight loss:
A B C D E F G
11 ANOVA
12 Source of Variation SS df MS F P-value F crit
13 Between Groups 189.4 3 63.14 1.78 0.1532 2.66
14 Within Groups 5533.6 156 35.47
15
16 Total 5723.0 159
F = 1.78, p-value = .1532. There is not enough evidence to conclude that there are differences in weight loss
between the four diets.

Percent LDL decrease:

A B C D E F G
11 ANOVA
12 Source of Variation SS df MS F P-value F crit
13 Between Groups 2569.4 3 856.48 32.56 0.0000 2.66
14 Within Groups 4104.0 156 26.31
15
16 Total 6673.4 159
F = 32.56, p-value = 0. There is enough evidence to conclude that there are differences in bad cholesterol reduction
between the four diets.

Percent HDL Increase:

A B C D E F G
11 ANOVA
12 Source of Variation SS df MS F P-value F crit
13 Between Groups 5595.0 3 1864.99 92.62 0.0000 2.66
14 Within Groups 3141.3 156 20.14
15
16 Total 8736.2 159
F = 96.62, p-value = 0. There is enough evidence to conclude that there are differences in good cholesterol increase
between the four diets.

© 2015 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a
license distributed with a certain product or service or otherwise on a password-protected website for classroom use.