Question1:

ONE WAY ANOVA

Research objective: Is there a significance difference in preference of travel due to different
frequency of travel.

Statistical Hypothesis:
H0: Maintained Hypothesis: There is no significance difference in preference of travel due to
different frequency of travel.
1 = 2

H1: Action Hypothesis: There is significance difference in preference of travel due to different
frequency of travel.
1  2

Technique Used: Anova Single Factor

Result:

SUMMARY
Groups Count Sum Average Variance
Light 10 46 4.6 3.6
0.76666666
Medium 10 41 4.1 7
1.06666666
Heavy 10 72 7.2 7

ANOVA
Source of
Variation SS df MS F P-value F crit
15.2944785 3.62102E 3.35413082
Between Groups 55.4 2 27.7 3 -05 9
1.81111111
Within Groups 48.9 27 1

Total 104.3 29

Interpretation:
Using fcal vs fcrit
Since fcal > fcrit (15.29>3.35), we reject the NULL/ Maintained Hypothesis.

Hence, we conclude that there is a significance difference in preference of travel due to

different frequency of travel.

TWO WAY ANOVA

Research objective:
1) Is there a significance difference in preference of travel due to different gender.
2) Is there a significance difference in preference due to different frequency of travel
 3) Is there a significance difference in preference due to interaction of frequency of travel
and gender.

Statistical Hypothesis:
H0: Maintained Hypothesis:
1 = 2 = 3

H1: Action Hypothesis:

1  2  3

Technique Used: Two way Anova

Results:

Light Medium Heavy

Male 3 4 8
Female 6 4 7

Chart Title
9
8
7
6
5
4
3
2
1
0
Light Medium Heavy

Male Female

Anova: Two-Factor With Replication

SUMMARY Light Medium Heavy Total

Male
Count 5 5 5 15
Sum 15 20 39 74
Average 3 4 7.8 4.933333333
Variance 1 1 0.7 5.352380952

Female
Count 5 5 5 15
Sum 31 21 33 85
Average 6.2 4.2 6.6 5.666666667
Variance 0.7 0.7 0.8 1.80952381

Total
 Count 10 10 10
Sum 46 41 72
Average 4.6 4.1 7.2
Variance 3.6 0.766666667 1.066666667

ANOVA
Source of
Variation SS df MS F P-value F crit
Sample 4.033333333 1 4.033333333 4.93877551 0.035939505 4.259677273
Columns 55.4 2 27.7 33.91836735 1.0147E-07 3.402826105
Interaction 25.26666667 2 12.63333333 15.46938776 4.83045E-05 3.402826105
Within 19.6 24 0.816666667

Total 104.3 29

Interpretation:
Using fcal vs fcrit
Since fcal > fcrit (15.469>3.402), we reject the NULL/ Maintained Hypothesis.

Hence, we conclude that there is a significance difference in preference of travel due to

interaction between different frequency of travel and gender.

Question2:

PROPORTION ANALYSIS:
Research objective: Is there a significance difference in satisfaction level from 2012 to 2017.

Statistical Hypothesis:
H0: Maintained Hypothesis: There is no significance difference in satisfaction level from 2012
to 2017.
p1 = p2

H1: Action Hypothesis: There is significance difference in satisfaction level from 2012 to 2017.
p1  p2

Technique Used: Hypothesis Test for two independent proportion.

Results:
Hypothesis test for two independent proportions

p1 p2 pc
0.3942 0.352 0.3712 p (as decimal)
41/104 44/125 85/229 p (as fraction)
41. 44. 85. X
104 125 229 n

0.0422 difference
0. hypothesized difference
 0.0641 std. error
0.66 z
.5101 p-value (two-tailed)

Interpretation:
Since p> (0.5101 > 0.05), we accept the null hypothesis.

Hence we conclude that there is no significance different in satisfaction level from 2012 to
2017.

Question3:

REGRESSION ANALYSIS:
Research objective: Is the change in relative sales explained by change in relative promotional
expenses.

Statistical Hypothesis:
H0: Maintained Hypothesis: There is no change in relative sales due to change in promotional
expense.
1 = 0

H1: Action Hypothesis: There is change in relative sales due to change in promotional expense.
1  0

Technique Used: Regression Testing

Results:
 Relative Sales - Y
140 y = 1.1488x - 7.9268
R² = 0.9871
120

100

80
Axis Title

Relative Sales - Y
60
Linear (Relative Sales - Y)
40

20

0
0 20 40 60 80 100 120 140
Axis Title

SUMMARY
OUTPUT

Regression Statistics
0.9935
Multiple R 11162
0.9870
R Square 64428
Adjusted R 0.9860
Square 69384
1.9573
Standard Error 75126
Observations 15

ANOVA
Signifi
cance
df SS MS F F
991.9
3800.5 3800.5 80693 1.1632
Regression 1 92874 92874 1 7E-13
49.807 3.8313
Residual 13 12599 17384
Total 14 3850.4
 Standa
Coeffic rd P- Lower Upper Lower Upper
ients Error t Stat value 95% 95% 95.0% 95.0%
- - 0.046 - - - -
7.9267 3.5957 2.2044 11599 15.694 0.1586 15.694 0.1586
Intercept 95388 58132 85145 2 95855 32224 95855 32224
Relative 1.163
Promotional 1.1488 0.0364 31.495 27E- 1.0700 1.2276 1.0700 1.2276
Expense -X 40117 76065 725 13 38369 41865 38369 41865

Interpretation:

Since p< (0.000 < 0.05), we reject the null hypothesis.

Since R2 = 0.987064428, therefore 98% change in relative sales is explained by change in

relative promotional expenses while the remaining 2% is explained by other factors.

Regression equation:
Y =  0 +  1X
0= -7.926795388
1= 1.148840117

y = 1.1488x - 7.9268

For different values of promotional expenses, we have calculated sales using the regression
equation.

Promotional expense-X 130 102 89

Sales-Y 141.4224198 109.2548965 94.319975

Question4:

PROPORTION ANALYSIS:
Research objective: To estimate the number of American adults who 50 year and older who
are concerned about health because of illness.

Statistical Hypothesis:
H0: Maintained Hypothesis:
p=0

H1: Action Hypothesis:

p0

Technique Used: Confidence Interval - p

Results:
Count (3) 236
Sample Size 1220
p 0.193443

For the sample, p=0.19. That means 20% of American adults are concerned about health
because of illness.

Population Size: 1130000

CI LCL UCL
95% 0.171 0.216
Population Estimate 193230 244080

We are 95% confident that estimated 1,93,230-2,44,080 American adults are concerned about
health because of illness.