Sei sulla pagina 1di 12

(a) Run the descriptive statistics on the data.

Annual Travel Expenses to Japan


No LI (RM) AC (RM) BA (RM)
1 7500 6000 9000
2 8000 8000 8450
3 5000 7000 8500
4 4000 7500 8800
5 6500 6500 8850
6 7000 5500 8500
7 5000 7000 9300
8 7000 6000 7800
9 7000 6000 9100
10 6500 6500 8200
Total 63500 66000 86500
Mean 6350 6600 8650
Standard Deviation 1270.3893 774.5967 449.6913

Note
LI = Logistics and international shipping program
AC = Accounting program
BA = Business administration program

(i) Highlight the sample means with yellow colour and interpret its meanings.
Interpretation- mean:
The average spends on travelling to Japan by 10 logistics and international shipping
program students is RM6350.
The average spends on travelling to Japan by 10 accounting program students is RM6600.
The average spends on travelling to Japan by 10 business administration program students
is RM8650.

(ii) Underline the standard deviations and interpret its meanings.


Interpretation- Standard Deviation
The standard deviation for logistics and international shipping program is the highest
among the three programs, which is RM 1270.39. This means that the values in the sample set
for logistics and international shipping program are the furthest from the mean, which is
RM6350. The amount spent by the students from logistics and international shipping program
is in between RM5079.61 and RM7620.39.

1
The amount spent by the students from accounting program is in between RM5825.40 and
RM7374.60.
The standard deviation for business administration program is the lowest among the three
programs which is RM 449.69. This means that the values in the sample set for business
administration program are the closest to the mean, which is RM8650. The amount spent by
the students from business administration program is in between RM8200.31 and RM9099.69.

(iii) Graphically present the data (histogram) and compare among the programs.

Annual Travel Expenses to Japan


among LI, AC and BA Students
8
Number of Students

6
4
2
0
4000-5000 5000-6000 6000-7000 7000-8000 8000-9000 9000-10000
Annual Travel Expenses (RM)

LI AC BA

Average Annual Travel Expenses to Japan


among LI, AC and BA Students
Average Annual Travel Expenses

10000 8650
8000
6350 6600
6000
(RM)

4000

2000

0
LI AC BA
Course

From the histogram generated above, we can conclude that the business administration
students will spend the most to travel to Japan. Accounting students are the second while
logistics and international shipping program will spend the least to travel to Japan. The
difference between the highest and the lowest average annual travel expenses is RM 2300.

2
2(b) Based on the results obtained from EXCEL output, perform the hypothesis using the
different approaches at 𝜶 = 0.05.
(i) Run an ANOVA analysis to verify the statistical difference in the mean of the annual
spending among the 3 chosen programs.

ANOVA: Single Factor


Summary
Program Count Sum Average Variance
Logistics and International Shipping 10 63500 6350 1613889
Accounting 10 66000 6600 600000
Business Administration 10 86500 8650 202222.2

ANOVA

Source of Variation SS df MS F P-value F crit


Between Groups 31850000 2 15925000 19.77351 5.14E-06 3.354131
Within Groups 21745000 27 805370.4

Total 53595000 29

ANOVA 7 Steps Hypothesis


1. Critical Value Approach
Let μ1 = Logistics and International Shipping
μ2 = Accounting
μ3 = Business Administration

Step 1 H0: μ1 = μ2 = μ3
H1: At least one of μi is not equal; where i = 1, 2, 3 (1=LI, 2=AC, 3=BA)
Step 2 α = 0.05
Step 3 Decision rule: Reject H0 if F statistic > Fc, otherwise do not reject H0.
Step 4 Critical value, Fc = 3.354131
Step 5 F statistic = 19.77351
Step 6 Decision Making: Reject H0 at α = 0.05, since F statistics = 19.77351 > 3.354131.

3
Step 7 Conclusion: Sufficient evidence to conclude that there are differences in the annual
expenses on travelling to Japan among Logistics and International
Shipping, Accounting, and Business Administration students whereby at
least one of the UTAR population mean by programs µi differs from the
rest at α = 0.05.

2. P-value Approach

Step 1 H0: μ1 = μ2 = μ3
H1: At least one of μi is different, where i = 1, 2, 3 (1=LI, 2=AC, 3=BA)
Step 2 α = 0.05
Step 3 Decision rule: Reject H0 if p-value < α, otherwise do not reject H0.
Step 4 F statistic = 19.77351
Step 5 From the excel data, p-value = 0.0000051425 when F = 19.77351
Step 6 Decision Making: Reject H0 since the p-value = 0.0000051425 < significance level,
α = 0.05.
Step 7 Conclusion: Sufficient evidence to conclude that there are differences in the annual
expenses on travelling to Japan among Logistics and International
Shipping, Accounting, and Business Administration students whereby at
least one of the UTAR population mean by programs µi differs from the
rest at α = 0.05.

4
(ii) Run two independent samples mean t-test with unequal variances to verify the
comparison in the performance of the variable between 2 programs.
t-Test: Two-Sample Assuming Unequal Variances
i) LI & AC

LI AC
Mean 6350 6600
Variance 1613888.889 600000
Observations 10 10
Hypothesized Mean Difference 0
df 15
t Stat -0.53132726
P(T<=t) one-tail 0.301485935
t Critical one-tail 1.753050356
P(T<=t) two-tail 0.602971871
t Critical two-tail 2.131449546

Critical value approach


Let µ1 = LI students and µ2 = AC students
H0: µ1- µ2 = 0
H1: µ1- µ2 ≠ 0

Level of significance, α = 0.05

Decision Rule: Reject H0 when the test statistic value is lower than lower bound critical value
or greater than upper bound critical value. Otherwise, do not reject H0.
Critical value, t: ± 2.1314

Degree of freedom: 15
(𝑋̅1 −𝑋̅2 )−(𝜇1 −𝜇2 )
Test statistic: t =
𝑆 2
𝑆 2
√ 1+ 2
𝑛1 𝑛2

(6350−6600)−(0−0)
= 1613888.889 600000
√ +
10 10

= -0.5313

-2.1314 -0.5313 2.1314

5
Decision Making: Do not reject H0 since the test statistic -0.5313 is greater than the lower
critical value -2.1314 and lower than upper critical value 2.1314.
Conclusion: There is insufficient evidence to conclude that there is difference in average
annual expenses on travelling to Japan between LI students and AC students at α
= 0.05.

P-value Approach
Let µ1 = LI students and µ2 = AC students
H0: µ1- µ2 = 0
H1: µ1- µ2 ≠ 0

Level of significance, α = 0.05

Decision Rule: Reject H0 if P-value is less than 𝛼 (0.05). Otherwise, do not reject H0.
P-Value: 0.6030

Decision Making: Do not reject H0 because the P-value is 0.6030 which is greater than 𝛼=0.05.
Conclusion: There is insufficient evidence to conclude that there is difference in average
annual expenses on travelling to Japan between LI students and AC students at α
= 0.05.

Confidence Interval (CI) Approach


Let µ1 = LI students and µ2 = AC students
H0: µ1= µ2
H1: µ1≠µ2

Level of significance, 𝜶 = 0.05


Decision Rule: Reject H0 if the hypothesized average annual expenses difference is not
included in the confidence interval. Otherwise, do not reject H0.
Critical Value = ±2.1314
95% of Confidence Interval
𝑆2 𝑆2
CI = (𝑋̅1 − 𝑋̅2 ) ± 𝑡𝛼,𝑣 √(𝑛1 + 𝑛2 )
2 1 2

1613888.889 600000
= (6350-6600)± 2.1314√( + )
10 10

= (-1252.8659, 752.8659)

6
95%

-1252.8659 752.8659

Decision Making: Do not reject H0 because the hypothesized average annual expenses
difference is within the confidence interval (-1252.8659, 752.8659).
Conclusion: There is insufficient evidence to conclude that there is difference in average
annual expenses on travelling to Japan between LI students and AC students.
Interpretation: We are 95% confident to conclude that the difference in average annual
expenses on travelling to Japan between LI students and AC students is
between -1252.8659 and 752.8659.

ii) LI & BA
LI BA
Mean 6350 8650
Variance 1613888.889 202222.2222
Observations 10 10
Hypothesized Mean Difference 0
df 11
t Stat -5.397052321
P(T<=t) one-tail 0.000108816
t Critical one-tail 1.795884819
P(T<=t) two-tail 0.000217633
t Critical two-tail 2.20098516

Critical value approach


Let µ1 = LI students and µ2 = BA students
H0: µ1- µ2 = 0
H1: µ1- µ2 ≠ 0

Level of significance, α = 0.05

Decision Rule: Reject H0 when the test statistic value is lower than lower bound critical value
or greater than upper bound critical value. Otherwise, do not reject H0.
Critical value, t: ± 2.2010

Degree of freedom: 11

7
(𝑋̅1 −𝑋̅2 )−(𝜇1 −𝜇2 )
Test statistic: t =
𝑆 2
𝑆 2
√ 1+ 2
𝑛1 𝑛2

(6350−8650)−(0−0)
= 1613888.889 202222.2222
√ +
10 10

= -5.3971

-2.2010 2.2010
-5.3971

Decision Making: Reject H0 since the test statistic -5.3971 is smaller than the lower critical
value -2.2010.
Conclusion: There is sufficient evidence to conclude that there is difference in average annual
expenses on travelling to Japan between LI students and BA students at α = 0.05.

P-value Approach
Let µ1 = LI students and µ2 = BA students
H0: µ1- µ2 = 0
H1: µ1- µ2 ≠ 0

Level of significance, α = 0.05

Decision Rule: Reject H0 if P-value is less than 𝛼 (0.05). Otherwise, do not reject H0.
P-Value: 0.0002176

Decision Making: Reject H0 because the P-value is 0.0002176 which is smaller than 𝛼=0.05.
Conclusion: There is sufficient evidence to conclude that there is difference in average annual
expenses on travelling to Japan between LI students and BA students at α = 0.05.

Confidence Interval (CI) Approach


Let µ1 = LI students and µ2 = BA students
H0: µ1= µ2
H1: µ1≠µ2

Level of significance, 𝜶 = 0.05

8
Decision Rule: Reject H0 if the hypothesized average annual expenses difference is not
included in the confidence interval. Otherwise, do not reject H0.
Critical Value = ±2.2010
95% of Confidence Interval
𝑆2 𝑆2
CI = (𝑋̅1 − 𝑋̅2 ) ± 𝑡𝛼,𝑣 √(𝑛1 + 𝑛2 )
2 1 2

1613888.889 202222.2222
= (6350-8650)± 2.2010√( + )
10 10

= (-3237.9750, -1362.0250)

95%

-3237.9750 -1362.0250

Decision Making: Reject H0 because the hypothesized average annual expenses difference is
out of the confidence interval (-3237.9750, -1362.0250).
Conclusion: There is sufficient evidence to conclude that there is difference in average annual
expenses on travelling to Japan between LI students and BA students.
Interpretation: We are 95% confident to conclude that the difference in average annual
expenses on travelling to Japan between LI students and BA students is not
between -3237.9750 and -1362.0250.

iii) AC & BA

AC BA
Mean 6600 8650
Variance 600000 202222.2222
Observations 10 10
Hypothesized Mean
Difference 0
df 14
t Stat -7.23779898
P(T<=t) one-tail 0.000002152
t Critical one-tail 1.761310136
P(T<=t) two-tail 0.000004305
t Critical two-tail 2.144786688

9
Critical value approach
Let µ1 = AC students and µ2 = BA students
H0: µ1- µ2 = 0
H1: µ1- µ2 ≠ 0

Level of significance, α = 0.05

Decision Rule: Reject H0 when the test statistic value is lower than lower bound critical value
or greater than upper bound critical value. Otherwise, do not reject H0.
Critical value, t: ± 2.1448

Degree of freedom: 14
(𝑋̅1 −𝑋̅2 )−(𝜇1 −𝜇2 )
Test statistic: t =
𝑆 2
𝑆 2
√ 1+ 2
𝑛1 𝑛2

(6600−8650)−(0−0)
= 600000 202222.2222
√ +
10 10

= -7.2378

-2.1448 2.1448
-7.2378

Decision Making: Reject H0 since the test statistic -7.2378 is smaller than the lower critical
value -2.1448.
Conclusion: There is sufficient evidence to conclude that there is difference in average annual
expenses on travelling to Japan between AC students and BA students at α = 0.05.

P-value Approach
Let µ1 = AC students and µ2 = BA students
H0: µ1- µ2 = 0
H1: µ1- µ2 ≠ 0

Level of significance, α = 0.05

Decision Rule: Reject H0 if P-value is less than 𝛼 (0.05). Otherwise, do not reject H0.

10
P-Value: 0.000004305

Decision Making: Reject H0 because the P-value is 0.000004305 which is smaller than 𝛼=0.05.
Conclusion: There is sufficient evidence to conclude that there is difference in average annual
expenses on travelling to Japan between AC students and BA students at α = 0.05.

Confidence Interval (CI) Approach


Let µ1 = AC students and µ2 = BA students
H0: µ1= µ2
H1: µ1≠µ2

Level of significance, 𝜶 = 0.05


Decision Rule: Reject H0 if the hypothesized average annual expenses difference is not
included in the confidence interval. Otherwise, do not reject H0.
Critical Value = ±2.1448
95% of Confidence Interval
𝑆 2𝑆 2
CI = (𝑋̅1 − 𝑋̅2 ) ± 𝑡𝛼,𝑣 √(𝑛1 + 𝑛2 )
2 1 2

600000 202222.2222
= (6600-8650)± 2.1448√( + )
10 10

= (-2657.4830, -1442.5170)

95%

-2657.4830 -1442.5170

Decision Making: Reject H0 because the hypothesized average annual expenses difference is
out of the confidence interval (-2657.4830, -1442.5170).
Conclusion: There is sufficient evidence to conclude that there is difference in average annual
expenses on travelling to Japan between AC students and BA students.
Interpretation: We are 95% confident to conclude that the difference in average annual
expenses on travelling to Japan between AC students and BA students is not
between -2657.4830 and -1442.5170.

11
(iii) Based on part 5(b)(i) & 5(b)(ii), draw an overall conclusion.
To analyze the average travel expenses among three different programs which are Business
Administration (BA), Accounting (AC) and Logistics and International Shipping (LI), we have
used ANOVA test, and T-test with 3 approaches including critical value approach, P-value
approach and confidence interval approach.
First of all, the result of ANOVA analysis and P-value approach have sufficient evidence
to reject H0. It is because the test statistic value (19.77351) is greater than the upper critical
value of 3.354131. Then the P-value approach also shows that the P-value (0.0000051425) is
less than the significance value 0.05. Therefore, we can conclude that the mean of Business
Administration, Accounting and Logistics and International Shipping have significance
difference in the average travel expenses to Japan at the significance level of 5%.
Next, according to T-test, for the comparison of Accounting with Logistics and
International Shipping, we have insufficient evidence to conclude that there is difference in
average annual expenses on travelling to Japan between LI students and AC students at α =
0.05. This is resulted in the three approaches: Critical value, the test statistics (-0.5313) is
greater than the lower critical value -2.1314 and lower than upper critical value 2.1314; P-value,
p-value (0.6030) is greater than 5% significant level; Confidence interval, the hypothesized
mean difference (0) falls between the confidence interval (-1252.8659, 752.8659).
Secondly, for the comparison of Business Administration with Logistic and International
Shipping, we have sufficient evidence to conclude that there is difference in average annual
expenses on travelling to Japan between LI students and BA students at 5% significance level.
This is resulted in the three approaches: Critical value, the test statistics (-5.3971) is smaller
than the lower critical value of -2.2010; P-value, p-value (0.0002176) is smaller than alpha
(0.05); Confidence interval, the hypothesized mean difference (0) falls out of the confidence
interval (-3237.9750, -1362.0250).
Lastly, for the comparison of Business Administration with Accounting, we have sufficient
evidence to conclude that there is difference in average annual expenses on travelling to Japan
between AC students and BA students at 0.05 significant level. This is resulted in the three
approaches: Critical value, the test statistics (-7.2378) is in smaller than the lower critical value
-2.1448; P-value, p-value (0.000004305) is smaller than alpha (0.05); Confidence interval, the
hypothesized mean difference (0) falls out of the confidence interval (-2657.4830, -1442.5170).
As a result, we can conclude that there is no difference in the average travel expenses to
Japan between Accounting with Logistics and International Shipping students while there is a
difference in that of between Business Administration with Accounting students, and Business
Administration with Logistic and International Shipping students at 95% significance level.

12

Potrebbero piacerti anche