Example 9.
9 X
According to HowtoAdvice.com, the average price charged to a customer to have a 52
12 by 18 wall-to-wall carpet shampoo cleaned is about $50. Suppose that a start-up 52
carpet-cleaning company believes that in the region in which they operate, the 56
average price for this service is higher. To test this hypothesis, the carpet-cleaning 50
company randomly contacts 23 customers who have recently had a 12 by 18 50
wall-to-wall carpet shampoo cleaned and asked the customers how much they 51
were charged for the job. Suppose the resulting data are given below and that the 49
population standard deviation price is $3.49. Use a 10% level of significance to test 49
their hypothesis. Assume that such prices are normally distributed in the population. 54
What is the observed value? What is the p-value? What is the decision? If the null 51
hypothesis is rejected, is the result substantive? 51
sigma ^2 12.1801 48
56
52
52
53
H0 <= 50 56
H1 >50 52
SINCE sample size is less than 30 we use t test 52
56
57
48
53
52.17391
Example 9.9 X
According to HowtoAdvice.com, the average price charged to a customer to have a 52
12 by 18 wall-to-wall carpet shampoo cleaned is about $50. Suppose that a start-up 52
carpet-cleaning company believes that in the region in which they operate, the 56
average price for this service is higher. To test this hypothesis, the carpet-cleaning 50
company randomly contacts 23 customers who have recently had a 12 by 18 50
wall-to-wall carpet shampoo cleaned and asked the customers how much they 51
were charged for the job. Suppose the resulting data are given below and that the 49
population standard deviation price is $3.49. Use a 10% level of significance to test 49
their hypothesis. Assume that such prices are normally distributed in the population. 54
What is the observed value? What is the p-value? What is the decision? If the null 51
hypothesis is rejected, is the result substantive? 51
sigma ^2 12.1801 48
H0 µ= 50 56
H1 µ> 50 52
52
53
56
52
sanple size is less than 30 so we will use t test 52
more than 30 sample size, z test 56
57
48
53
52.17391
Dummy
0 t-Test: Paired Two Sample for Means
0
0 t-Test: Paired Two Sample for Means
0
0 X Dummy
0 Mean 52.17391 0
0 Variance 7.059289 0
0 Observations 23 23
0 Pearson Correlation #DIV/0!
0 Hypothesized Mean Difference 50
0 df 22
0 t Stat 3.923969
0 P(T<=t) one-tail 0.000363
0 t Critical one-tail 1.321237 alpha
0 P(T<=t) two-tail 0.000726
0 t Critical two-tail 1.717144
0
0 it is right tailed test since h1 is more than 50
0 as tstat is greater than t critical, tsta lies in the rejection region
0 the alternate hypothesis is supported
0 p value is less than alpha so therefore we reject the null hypothesis.
0
0
Example 9.19
Based on population figures and other general information on the U.S. population,
suppose it has been estimated that, on average, a family of four in the United States
spends about $1135 annually on dental expenditures. Suppose further that a regional
dental association wants to test to determine if this figure is accurate for their area of
the country. To test this, 22 families of four are randomly selected from the population
in that area of the country and a log is kept of the family’s dental expenditures
for one year. The resulting data are given below. Assuming that dental expenditures
are normally distributed in the population, use the data and an alpha of .05 to test
the dental association’s hypothesis.
H0 µ= 1135
H1 µ≠ 1135 two tail
t tetst the sample size
h0 mu = 1135
h1 mu =/= not equal to 1135 two tail
equa
X dummy
1008 0
812 0
1117
1323
1308
1415
831
1021
1287
851
930
730
699
872
913
944
954
987
1695 t-Test: Two-Sample Assuming Unequal Variances
995
1003 X dummy
994 Mean 1031.31818 0
Variance 57779.9416 0
Observations 22 2
Hypothesized Mean Differe 1135
df 21
t Stat -2.0231375
P(T<=t) one-tail 0.0279869
t Critical one-tail 1.7207429
P(T<=t) two-tail 0.0559738
t Critical two-tail 2.0796138
tstat is less than t critical, t stat lies in the rejection
EX 10.8
Suppose a market analyst wants to determine the difference in the average price of a Seattle X1
gallon of whole milk in Seattle and Atlanta. To do so, he takes a telephone survey of 2.55
21 randomly selected consumers in Seattle who have purchased a gallon of milk 2.67
and asks how much they paid for it. The analyst undertakes a similar survey in 2.5
Atlanta with 18 respondents. Assume the population variance for Seattle is 0.03, 2.61
the population variance for Atlanta is 0.015, and that the price of milk is normally 2.43
distributed. Using the resulting sample information that follows 2.36
2.5
Using a 1% level of significance, test to determine if there is a significant difference 2.36
in the price of a gallon of milk between the two cities 2.54
2.54
2.8
Since sample size is greater than 30 we will use z test 2.61
2.56
Seattle Atlanta 2.64
Sample 21 18 2.43
2.43
H0 Mu1=mu2 2.38
h1 Mu1 not equal to mu2 2.49
2.57
2.71
2.27
Atlanta X2
2.25
2.3
2.49
2.41
2.39
2.26
2.4
2.33
2.29
2.48
2.59
2.38
2.39
2.4 z-Test: Two Sample for Means
2.23
2.29 Seattle X1 Atlanta X2
2.53 Mean 2.5214285714 2.381111
2.45 Known Variance 0.03 0.015
Observations 21 18
Hypothesized Mean Difference 0
z 2.9503569207
P(Z<=z) one-tail 0.0015870351
z Critical one-tail 2.326347874
P(Z<=z) two-tail 0.0031740702
z Critical two-tail 2.5758293035
Hypothesis
mu 1= mu2
mu1 not equal to mu2 two tail
z stat lies in the rejection region
p value is less than alpha w reject h0
there is a significant diff in the avg price sof eattle an atlanta
Ex 10.20
Some studies have shown that in the United States, men spend more than women
buying gifts and cards on Valentine’s Day. Suppose a researcher wants to test
this hypothesis by randomly sampling nine men and 10 women with comparable
demographic characteristics from various large cities across the United States to
be in a study. Each study participant is asked to keep a log beginning one month
before Valentine’s Day and record all purchases made for Valentine’s Day during
that one-month period. The resulting data are shown below. Use these data and
a 1% level of significance to test to determine if, on average, men actually do
spend significantly more than women on Valentine’s Day. Assume that such
spending is normally distributed in the population and that the population
variances are equal
t test since sample is 19
ho mu1 <= mu2, I,e u1-u2 =0
h1 mu1>mu2 u1-u2>0
m
Rejection region
Men X1 Women X2
107.48 125.98
143.61 45.53
90.19 56.35
125.53 80.62
70.79 46.37
83 44.34
129.63 75.21
154.22 68.48
93.8 85.84
126.11
t-Test: Two-Sample Assuming Equal Variances
Men X1 Women X2
Mean 110.9167 75.483
Variance 828.9558 930.7244
Observations 9 10
Pooled Variance 882.8333
Hypothesized Mean Difference 0
df 17
t Stat 2.595501
P(T<=t) one-tail 0.009427
t Critical one-tail 2.566934
P(T<=t) two-tail 0.018855
t Critical two-tail 2.898231
it is a right tailed test
p value < alpha
t stat is greater than t critical so t stat lies in the rejection region
h0 is rejected
Women X2
t-Test: Two-Sample Assuming Equal Variances
Men X1 Women X2
Mean 110.9167 75.483
Variance 828.9558 930.7244
Observatio 9 10
Pooled Var 882.8333
Hypothesiz 0
df 17
t Stat 2.595501
P(T<=t) one 0.009427
t Critical on 2.566934
P(T<=t) two 0.018855
t Critical tw 2.898231
Hypothesis
H0 : µ1 = µ2 i.e µ1 - µ2 =0
H1: µ1 > µ2 i.e µ1 - µ2 >0 right tail
Rejection region
t> 2.566934
Test Statistic
Tstat =2.595501
Tstat is in rejection region
pvalue = 0.009427
alpha= 0.01
p value < alpha
Decision
Tstat is in rejection region and p value less than alpha so we rehect H0
There is enough evidence to infer that men spend significantly more than women for Valentines day
EX 10.65
Employee
In manufacturing, does worker productivity drop on 1
Friday? In an effort to determine whether it does, a 2
company’s personnel analyst randomly selects from a 3
manufacturing plant five workers who make the same 4
part. He measures their output on Wednesday and 5
again on Friday and obtains the following results.
The analyst uses alpha = 0.05 and assumes the difference in
productivity is normally distributed. Do the samples
provide enough evidence to show that productivity
drops on Friday?
Ho mu 1= < m2
h1 mu 1 > mu2
t test
Friday
Wednesday Output X1 Output X2
71 53
56 47
75 52
68 55
74 58
t-Test: Paired Two Sample for Means
Wednesday Output X1 Friday Output X2
Mean 68.8 53
Variance 58.7 16.5
Observations 5 5
Pearson Correlation 0.763137142
Hypothesized Mean Difference 0
df 4
t Stat 6.712776761
P(T<=t) one-tail 0.001281871
t Critical one-tail 2.131846786
P(T<=t) two-tail 0.002563741
t Critical two-tail 2.776445105
Ho mu 1= < m2, i.e. mu1-mu2 =0 or less than 0
h1 mu 1 > mu2
it Is a right tailed test
p value < alpha
t stat is more than the t critical; so ut lies inr ejection
we reject null hypthissi
18
t-Test: Paired Two Sample for Means 9
23
Wednesday Output
FridayX1
Output X2 13
Mean 68.8 53 16
Variance 58.7 16.5
Observatio 5 5
Pearson Cor0.763137
Hypothesiz 0
df 4
t Stat 6.712777
P(T<=t) one 0.001282
t Critical on 2.131847
P(T<=t) two 0.002564
t Critical tw 2.776445
Hypothesis
H0 : µ1 = µ2 i.e µ1 - µ2 =0
H1: µ1 > µ2 i.e µ1 - µ2 >0 right tail
Rejection region
t>2.131847
tstat =6.712777
tstat is in rejection region
p value=0.001282
Decision
tstat is in rejection region and p value is less than alpha
The null hypothesis is rejected
There is enough evidence to infer that productivity drops on Friday
Observed frequency (f0) Table Observed frequency (f0) Table
smoker non smoker smoker non smoke
male 29 71 male 29 71
female 16 84 female 16 84
Ct 45 155
H0: Smoking status is independent of gender
H1: Smoking status is not independent of gender
LOS alpha =.05
(fo-fe)^2/fe
smoker non smoker
male 1.877778 0.545161
HO: Smoking status indep female 1.877778 0.545161
Expected frequency (fe) table
Rt smoker non smoker
100 male 22.5 77.5
100 female 22.5 77.5
200
df=(R-1)(C-1) 1
non smoker
Chi-square calculated 4.84587813620072 Chi square table 3.841459
p value 0.0277122790251598 3.841459
Chi square stat > Chi square critical. Chi sqaure calculated is in rejection region
p value < alpha
We reject the null hypothesis
There is enough evidence to infer that the smoking staus is not independent of gender
non smoker
Observed frequency (f0) Table Observed frequency (f0) Table
smoker non smoker smoker non smoker
male 29 71 male 29 71 100
female 16 84 female 16 84 100
45 155
HO: Smoking status independent of gender
H1: Smoking status not independent of gender degree of freedom= (c-1) (d-1)
Expected frequency (f0) Table
smoker non smoker
male 22.5 77.5 100
female 22.5 77.5 100
45 155 200
Degree of freedom 1
fo fe fo-fe (fo-fe)^2 (fo-fe)^2/fe
29 22.5 6.5 42.25 1.87777778
71 77.5 -6.5 42.25 0.54516129
16 22.5 -6.5 42.25 1.87777778
84 77.5 6.5 42.25 0.54516129
Chi square 4.84587814
chi square table 3.84145882
chi squre stat> chi square critical so chi square lies in the rejection region
therefore, we can reject the null hypothesis.
and say that smokign status in indepenent of gender
re lies in the rejection region
Ex 11.13
A management consulting company presents a three-day seminar on project High Level
management to various clients. The seminar is basically the same each time it is 7
given. However, sometimes it is presented to high-level managers, sometimes to 7
midlevel managers, and sometimes to low-level managers. The seminar facilitators 8
believe evaluations of the seminar may vary with the audience. Suppose the 7
following data are some randomly selected evaluation scores from different levels of 9
managers who attended the seminar. The ratings are on a scale from 1 to 10, with 10
being the highest. Use a one-way ANOVA to determine whether there is a significant
difference in the evaluations according to manager level. Assume = .05. Discuss
the business implications of your findings.
Anova: Single Factor
SUMMARY
Groups Count Sum Average Variance
High Level 5 38 7.6 0.8
Midlevel 7 62 8.857143 0.809524
Low Level 6 35 5.833333 2.166667
ANOVA
Source of Variation SS df MS F P-value F crit
Between G 29.60952 2 14.80476 11.75573 0.000849 3.68232034
Within Gro 18.89048 15 1.259365
Total 48.5 17
Midlevel Low Level Anova: Single Factor
8 5
9 6 SUMMARY
8 5 Groups Count Sum Average Variance
10 7 High Level 5 38 7.6 0.8
9 4 Midlevel 7 62 8.857143 0.809524
10 8 Low Level 6 35 5.833333 2.166667
8
ANOVA
Source of Variation SS df MS F P-value
Between Groups 29.60952 2 14.80476 11.75573 0.000849
Within Groups 18.89048 15 1.259365
Total 48.5 17
F crit
3.68232
