You are on page 1of 17

BEO2255 APPLIED STATISTICS FOR BUSINESS

ASSIGNMENT 2

PART A
Question 1

Table 1: Descriptive Statistics for Income

N Minimum Maximum Mean Std. Deviation

Income 100 180 580 367.20 120.839

Valid N (listwise) 100

Table 1 shows that the mean in the weekly income of the respondent is $367.20 with a
standard deviation of $120.839.

Mean = 367.20

Standard Deviation = 120.84

Finite population:

1. 95% confidence interval for population mean


=Y + 1.96
1

120.839 3000100
= 367.20 + 1.96
100 30001

= 367.20 + 19.341

= ($347.859, $386.541)

Conclusion
We are 95% confident that the mean of the income of the respondents is between $347.859
and $386.541.
Question 2
2a.

Table 2a: Descriptive Statistics for Income - Male


N Minimum Maximum Mean Std. Deviation

Income 57 180 580 359.47 119.140


Valid N (listwise) 57

2b.

Table 2b: Descriptive Statistics for Income - Female


N Minimum Maximum Mean Std. Deviation

Gender 43 180 580 377.44 123.714


Valid N (listwise) 43

Table 2a show that the mean and standard deviation for male in income are $359.47 and $119.140.
For table 2b, it show that the mean and standard deviation for female are $377.44 and $123.714

a. The sample mean is calculated as below:


N1 1
=

1600(359.47)+1400(377.44)
= 3000
1103568
= 3000

= 367.856

b. The standard error is calculated as below


1 2
ST = 2 [ 11
1
]

1 160057 119.1402 140043 123.7142


= (1600)2 [ ] [ 57 ] + (1400)2 [ 14001 ] [ ]
3000 16001 43
1
= 3000 615173704.7 + 676686398.2
1
= 3000 (35942.455)

=11.981
c. The 95% confidence interval is calculated as below:

1.96
= 367.856 1.96 (11.981)
=367.856 13.941
= (353.915, 381.797)

Conclusion

We are 95 % confident that the mean income for the male respondents and female
respondents within the population is between 353.915 and 381.797.

Question 3

Table 4: Descriptive Statistics: Male

N Minimum Maximum Mean Std. Deviation

POSITIVE 57 0 1 .32 .469

Valid N (listwise) 57

Table 5 shows the proportion of Male respondents for the mean has a positive respond 0.32
with a standard deviation 0.469.

The 95% confidence in the interval of proportion positive of male respondent to attitude is
calculated as below:


1.96 1

0.32(10.32) 160057
= 0.32 1.96
57 16001

= 0.32 1.96 (0.062) (0.982)


= 0.32 0.119
= (0.201, 0.439)
Conclusion
We are 95% confident that the proportion of male customers in the population with positive
response is between 20.1% and 43.9%

Question 4A
Wilcoxon Rank Sum Test
Hypothesis

H0: The distributions of purchase in the month before being exposed to the advertising
campaign are the same as for Full-time workers and Part-time workers in the population of
customers
H1: The distributions of purchase in the month before being exposed to the advertising
campaign are not the same as for Full-time workers and Part-time workers in the population
of customers

Table 5: Descriptive Statistics


N Mean Std. Deviation Skewness Kurtosis

Statistic Statistic Statistic Statistic Std. Error Statistic Std. Error

Before 100 47.45 16.065 .202 .241 -1.004 .478


After 100 50.21 25.114 .730 .241 .706 .478
Valid N (listwise) 100
Table 6: Descriptive Statistic: Ranks

Job N Mean Rank Sum of Ranks

45 59.37 2671.50
Part time

55 43.25 2378.50
Before Full-time

100
Total
Table 7: Test Statisticsa

Before

Mann-Whitney U 838.500

Wilcoxon W 2378.500

Z -2.767

Asymp. Sig. (2-tailed) .006

a. Grouping Variable: Job


It can be observed that:
The significance of the Wilcoxon Rank Sum Test is 0.006, which is smaller than 0.05,
the level of significance for the test. Therefore, we reject the null hypothesis.
The significant value of 0.006 indicates there is relatively strong evidence from the
sample to reject the null hypothesis.

Conclusion
At the 5% level of significance, there is sufficient evidence to conclude that the
distributions of purchase in the month before being exposed to the advertising campaign
are not the same for Full-time workers and Part-time workers in the population of
customers.
Question 4B
Wilcoxon Signed Rank Test
Hypothesis

H0: The purchase in the month before the advertising exposure is the same as the month after
for the population of customers.

H1: The purchase in the month before the advertising exposure is not the same as the month
after for the population of customers.

Table 8: Ranks
Mean
N Sum of Ranks
Rank

Negative Ranks 181a 114.11 20654.00

Positive Ranks 113b 200.98 22711.00


AFTER - BEFORE
Ties 6c

Total 300

a. AFTER < BEFORE


b. AFTER > BEFORE
c. AFTER = BEFORE

Table 9: Test Statisticsa

AFTER - BEFORE

Z -.706b
Asymp. Sig. (2-tailed) .480

a. Wilcoxon Signed Ranks Test


b. Based on negative ranks.
It can be observed that:

The significance of the Wilcoxon Signed Rank Test is 0.48, which is larger than 0.05,
the level of significance for the test. Therefore, we do not reject the null hypothesis.
The significant value of 0.48 indicates that there is insufficient evidence from the
sample to reject the null hypothesis.
Conclusion

At the 5% level of significance, there is insufficient evidence to conclude that the


purchase in the month before the advertising exposure is not the same as the month
after for the population of customers.

Question 5A
Chi-Square Goodness of Fit Test
Hypothesis

H0: Positive, neutral and negative responses are equally likely among the population of part-time
workers.
H1: Positive, neutral and negative responses are not equally likely amongst the population of part-time
workers.

Table 10: SUCCESS


Observed N Expected N Residual

Positive 32 33.3 -1.3


Neutral 20 33.3 -13.3
Negative 48 33.3 14.7
Total 100

Table 11: Test Statistics


SUCCESS

Chi-Square 11.840a
df 2
Asymp. Sig. .003

a. 0 cells (0.0%) have expected frequencies less than 5. The minimum expected cell frequency is 33.3.
Table 10 shows the Chi Square Goodness of Fit Test of the population of part-time workers response towards the
advertisement campaign.

It can be observed that:


The significance of Chi-Square Goodness of fit test is 0.003, which is smaller than 0.05, the level of significance
for the test. Therefore, we reject the null hypothesis.
The significance value of 0.003 indicates that there is relatively strong evidence from the sample to reject the null
hypothesis.
Conclusion:
At 5% level of significance, there is adequate confirmation to conclude that the number of population of part-time workers
have no less than one contrast response toward the impact of the customer preferences towards the product range.
Case Processing Summary

Cases

Valid Missing Total

N Percent N Percent N Percent

SUCCESS * ADVERTISING 100 100.0% 0 0.0% 100 100.0%

Question 5B
Chi-Square Test of Independence
H0: Attitudes and the method of advertising campaign are independent.
H1: Attitudes and the method of advertising campaign are dependent.

Table 12: Chi-Square Tests

Value df Asymp. Sig. (2-sided)

Pearson Chi-Square 3.905a 6 .690

Likelihood Ratio 4.100 6 .663

Linear-by-Linear Association .022 1 .883

N of Valid Cases 100

a. 1 cells (8.3%) have expected count less than 5. The minimum expected count is 3.80.

It can be observed that:


The significance value is 0.690 which is higher than 0.05 which known as the level of
significance.
We do not reject null hypothesis, since the test found that attitudes and the method of
advertising campaign are independent.
Conclusion:
At 5% significance of level, there is sufficient evident to do not reject the null hypothesis that
attitude and success are independent.

Question 5C
Cross tabulate of SUCCESS with ADVERTISING methods

Table 13: SUCCESS * ADVERTISING Crosstabulation


ADVERTISING
Total
Email Leaflets Newspaper Social Media

Count 6 10 6 10 32

Expected Count 6.1 9.3 8.0 8.6 32.0


Positive % within
31.6% 34.5% 24.0% 37.0% 32.0%
ADVERTISING

% of Total 6.0% 10.0% 6.0% 10.0% 32.0%

Count 3 8 6 3 20

Expected Count 3.8 5.8 5.0 5.4 20.0


SUCCESS Neutral % within
15.8% 27.6% 24.0% 11.1% 20.0%
ADVERTISING
% of Total 3.0% 8.0% 6.0% 3.0% 20.0%

Count 10 11 13 14 48

Expected Count 9.1 13.9 12.0 13.0 48.0


Negative % within
52.6% 37.9% 52.0% 51.9% 48.0%
ADVERTISING

% of Total 10.0% 11.0% 13.0% 14.0% 48.0%


Count 19 29 25 27 100

Expected Count 19.0 29.0 25.0 27.0 100.0


Total % within
100.0% 100.0% 100.0% 100.0% 100.0%
ADVERTISING
% of Total 19.0% 29.0% 25.0% 27.0% 100.0%

It can be observed that:



PART B
Question 6
One-Way ANOVA
H0: The mean sales are the same in all three display locations.
H1: The mean sales are different in at least one according to different display locations.

Table 14: SALES by DISPLAY One-Way ANOVA


Sum of Squares df Mean Square F Sig.

Between Groups 260154887.656 2 130077443.828 8.185 .001


Within Groups 1541469091.344 97 15891433.931
Total 1801623979.000 99

It can be observed that:

The significance level of the on-way ANOVA test of sales by success is 0.001, which is
smaller than the level of significance for the test 0.05.
We reject null hypothesis that the mean sales are the same in all three displays locations.
Conclusion:
At 5% significance level, there is sufficient evidence to reject null hypothesis and to conclude that the
mean sales are different in at least one according to different displays locations.

Question 7
Tukeys Confident Interval by DISPLAY

Table 15: Tukey HSD of SALES Multiple Comparisons


Mean Difference 95% Confidence Interval
(I) DISPLAY (J) DISPLAY Std. Error Sig.
(I-J) Lower Bound Upper Bound

Middle 1385.415 988.675 .344 -967.85 3738.68


Lower
Upper 3970.852* 1007.658 .000 1572.40 6369.30
Lower -1385.415 988.675 .344 -3738.68 967.85
Middle
Upper 2585.437* 947.045 .020 331.26 4839.61
Lower -3970.852* 1007.658 .000 -6369.30 -1572.40
Upper
Middle -2585.437* 947.045 .020 -4839.61 -331.26

*. The mean difference is significant at the 0.05 level.

Hypothesis:
H0: The mean sales difference between pair of display location for sales is not significant.
H1: The mean sales difference between pair of display location for sales is significant.

It can be observed:
Base on the significant of the test, the pair Upper-Lower-Middle has significant different
mean sales.
The significance value of the Upper-Lower is 0.000, which is smaller than 0.05, the level of
significance for the test. Therefore, we reject the null hypothesis.
The significance value of the Upper-Middle is 0.020, which is smaller than 0.05, the level of
significance for the test. Therefore, we reject the null hypothesis.
The significance value of 0.020 < (0.05) indicates that there is relatively strong evidence
from the sample to reject the null hypothesis.
We are 95% confidence that the mean sales different between Upper-Lower display location
is between $6369.30 and $1572.40.
We are 95% confidence that the mean sales difference between Upper-Middle display
locations is between $4839.61 and $331.26.

Conclusion:
At 5% of significant level, there is sufficient evidence to conclude that the mean sales difference
between Upper-Lower location (p=0.000) and Upper-Middle location (p=0.020) for sales is
significant.

Question 8
Two-Way ANOVA

Table 16: Between-Subjects Factors


Value Label N

1 Brand A 19

2 Brand B 29
PRODUCT
3 Brand C 25

4 Brand D 27
1 Leaflet 18

PROMOTION 2 Posters 27

3 Newspapers 25
4 Brochures 30

(no need to interpret for above table)

Table 17: Two-way ANOVA Test of SALES Tests of Between Subjects Effects
Type III Sum of
Source df Mean Square F Sig.
Squares

Corrected Model 991482714.771a 14 70820193.912 7.430 .000

Intercept 5230072670.412 1 5230072670.412 548.739 .000

Product 266161645.884 3 88720548.628 9.309 .000


Promotion 76284365.779 3 25428121.926 2.668 .053
Product * Promotion 408958164.069 8 51119770.509 5.363 .000
Error 810141264.229 85 9531073.697
Total 8576724700.000 100
Corrected Total 1801623979.000 99

a. R Squared = .550 (Adjusted R Squared = .476)


Hypothesis 1:
H0: There are no interaction effect between product and promotion that affect sales.
H1: There are interaction effect between product and promotion that affect sales.

Observation 1:
The significance value of the two-way ANOVA test between product and promotion is 0.000,
which is smaller than 0.05, the level of significance for the test.
The significance value of 0.000 indicates that there is strong evidence to reject the null
hypothesis. Therefore, we reject the null hypothesis.
Conclusion 1:
Based on the sample of 100 salespersons, at 5% significance level, there is sufficient evidence to
conclude that there are interaction effect between product and promotion that affect sales.

Hypothesis 2:
H0: The mean sales for all products are the same.
H1: At least one different in mean sales according to the different products.

Observation 2:
The significance of the two-way ANOVA test for products is 0.000, which is smaller than the
0.05, the level of significance for the test.
The significance value of 0.000 indicates that there is strong evidence to reject the null
hypothesis. Therefore, we reject the null hypothesis.
Conclusion 2:
Based on the sample 100 salespersons, at 5% of significance level, there is sufficient evidence to
conclude that there are at least one difference in mean sales according to the different products.

Hypothesis 3:
H0: The mean sales are the same for all promotion methods.
H1: At least one difference in mean sales according to promotion methods.

Observation 3:
The significance of the two-way ANOVA test for promotion methods is 0.053, which is
greater than the 0.05, the level of significance for the test.
The significance value of 0.053 indicates that there is insufficient evidence to reject the null
hypothesis. Therefore, we do not reject the null hypothesis.
Conclusion 3:
Based on the sample of 100 salespersons at 5% significance of level, there is insufficient evidence to
conclude that at least one difference in mean sales according to promotion methods.

Question 9
Tukey CI of Sales

Table 18: Tukey HSD of SALES Multiple Comparisons by PRODUCT

Mean 95% Confidence Interval


(I) PRODUCT (J) PRODUCT Std. Error Sig.
Difference (I-J) Lower Bound Upper Bound

Brand B 4642.47* 911.204 .000 2254.57 7030.36

Brand A Brand C 3416.36* 939.616 .003 954.01 5878.71

Brand D 228.34 924.466 .995 -2194.31 2650.99

Brand A -4642.47* 911.204 .000 -7030.36 -2254.57


Brand B Brand C -1226.11 842.556 .469 -3434.11 981.88
Brand D -4414.13* 825.627 .000 -6577.76 -2250.49

Brand A -3416.36* 939.616 .003 -5878.71 -954.01


Brand C Brand B 1226.11 842.556 .469 -981.88 3434.11
Brand D -3188.01* 856.881 .002 -5433.55 -942.48
Brand A -228.34 924.466 .995 -2650.99 2194.31

Brand D Brand B 4414.13* 825.627 .000 2250.49 6577.76

Brand C 3188.01* 856.881 .002 942.48 5433.55


Based on observed means.
The error term is Mean Square(Error) = 9531073.697.
*. The mean difference is significant at the 0.05 level.

Table 19: Tukey HSD of SALES Multiple Comparisons by PROMOTION

(I) Mean 95% Confidence Interval


(J) PROMOTION Std. Error Sig.
PROMOTION Difference (I-J) Lower Bound Upper Bound

Posters -1607.96 939.418 .324 -4069.79 853.87

Leaflet Newspapers 886.29 954.331 .790 -1614.62 3387.20

Brochures -882.11 920.438 .773 -3294.20 1529.98

Leaflet 1607.96 939.418 .324 -853.87 4069.79


Posters Newspapers 2494.25* 856.881 .023 248.72 4739.79
Brochures 725.85 818.966 .812 -1420.32 2872.03

Leaflet -886.29 954.331 .790 -3387.20 1614.62


Newspapers Posters -2494.25* 856.881 .023 -4739.79 -248.72
Brochures -1768.40 836.030 .157 -3959.29 422.49

Leaflet 882.11 920.438 .773 -1529.98 3294.20

Brochures Posters -725.85 818.966 .812 -2872.03 1420.32

Newspapers 1768.40 836.030 .157 -422.49 3959.29

Based on observed means.


The error term is Mean Square(Error) = 9531073.697.
*. The mean difference is significant at the 0.05 level.

Question10
Kruskal-Wallis Test of SALES

Table 20: Ranks


DISPLAY N Mean Rank

Lower 29 61.48

SALES Middle 37 54.20


Upper 34 37.10
Total 100

(no need to interpret the above table)

Table 21: Test Statisticsa,b

SALES

Chi-Square 12.013
df 2
Asymp. Sig. .002

a. Kruskal Wallis Test


b. Grouping Variable: DISPLAY

Question 11
Check normality of SALES

Table 22: Descriptive Statistics


N Mean Std. Deviation Skewness Kurtosis

Statistic Statistic Statistic Statistic Std. Error Statistic Std. Error

SALES 100 8231.10 4265.937 1.259 .241 .774 .478


Valid N (listwise) 100