Tax Preferences and Lodgement Methods

This assignment examines the tax filing habits of Australians and international students. It investigates their preference for self-preparation or appointing tax agents, exploring correlations between age, income, deduction amounts, and chosen lodgement methods. The analysis involves statistical tables (6.4-6.6) presenting multiple R, R Square, ANOVA results, and regression coefficients to understand the factors influencing tax filing choices in Australia.

Running Head: BUS708 STATISTICS AND DATA ANALYSIS
BUS708 Statistics and Data Analysis
Name of the Student
Name of the University
Author Note

1BUS708 STATISTICS AND DATA ANALYSIS
Table of Contents
1.0 Introduction...........................................................................................................................................2
2.0 Analysis of Lodgement Method of ATO Data.........................................................................................2
3.0 Analysis of Lodgement Method of Students Data.................................................................................3
4.0 Age Range and Lodgement Method......................................................................................................5
5.0 Lodgement Method and Total Income..................................................................................................6
6.0 Total Income and Total Deduction based on Lodgement Methods.......................................................9
7.0 Conclusion...........................................................................................................................................10
References.................................................................................................................................................11

2BUS708 STATISTICS AND DATA ANALYSIS
1.0 Introduction
Lodging tax returns is the prime responsibility of each and every citizen of any nation. Taxes
should be payed to the government (Gallemore and Labro 2015). Despite of this, all the citizens are not
supposed to pay the income taxes. There is an income limit. Any individual earning more than the
income limit in a year is liable to pay the taxes (Bitler, Hoynes and Kuka 2016). Thus, on the occasion of
lodging the tax returns, an individual can opt for two different ways. The first way is to appoint a
registered tax agent and pay him or her to do the work for the peoson. The other way is to prepare the
lodgement file by himself or herself. This study is mainly based on Australia and the preference of the
Australian citizens for tax agents or self-preparers will be assessed in this study.
Two different types of data has been collected for conducting this study. The first data has been
obtained from the Australian Taxation Office (ATO) website. The data on tax lodgement for the year
2013-2014 has been obtained and a sample of 1000 data has been extracted from the original for the
purpose of this study. The data thus collected is a secondary data as it is obtained from a website. The
variables involved such as gender is a categorical variable, age range is a categorical variable and
lodgement method is a categorical variable. Other two variables involved such as total income and total
deduction are both quantitative variables. In table 1 below are the first five cases of the first dataset.
Table 1.1: First five cases of the ATO sample dataset
Gender age_range Lodgment_method Tot_inc_amt Tot_ded_amt
1 0 S 2389 0
1 6 S 4936 0
1 0 S 2462 0
1 9 A 29448 425
1 0 A 49665 684
Unlike the first dataset, the second dataset contains information about the preference of tax
lodgement methods by 150 international students studying in Australian Universities. This data is thus
primary data and the variable involved in this dataset is a categorical variable.
2.0 Analysis of Lodgement Method of ATO Data
Table 2.1 shows a summary of the number of people of Australia who prefer tax agents and who
prefer themselves for the lodgement of tax returns. It can be seen that 73.2 percent of the people prefer
tax agents. The proportion of preference for tax agents is shown diagrammatically in figure 2.1. It has
also been obtained from the analysis that 70 – 76 percent of the people of Australia prefer tax agents.
This can be claimed with 95 percent confidence.
Table 2.1: Summary of Lodgement Methods of ATO Dataset
Row Labels
Count of
Lodgment_method
Count of
Lodgment_method2
A 732 73.20%
S 268 26.80%

3BUS708 STATISTICS AND DATA ANALYSIS
Grand Total 1000 100.00%
73%
27%
Proportion of Lodgement Methods from the ATO Dataset
A
S
Figure 2.1: Pie chart showing proportion of Lodgement Methods for ATO Dataset
Table 2.2: Proportion of Tax Agents of ATO Dataset
Sample Size 1000
Count of Successes 732
Confidence Level 95%
Sample Proportion 0.732
z Value 1.9600
Standard Error of the Proportion 0.014006284
Margin of Error 0.0275
Calculations for Computation of Confidence Interval
Interval Lower Limit 70.45%
Interval Upper Limit 75.95%
3.0 Analysis of Lodgement Method of Students Data
Table 3.1 shows a summary of the number of international students who prefer tax agents and
who prefer themselves for the lodgement of tax returns. It can be seen that 74 percent of the students
selected prefer tax agents. The proportion of preference for tax agents is shown diagrammatically in
figure 3.1. It has also been obtained from the analysis that 71 – 85 percent of the international students
prefer tax agents. This can be claimed with 95 percent confidence.

4BUS708 STATISTICS AND DATA ANALYSIS
Table 3.1: Summary of Lodgement Methods of Students Dataset
Values
Lodgement Method Frequency Proportion
A 134 0.74
S 46 0.26
Grand Total 180 1
78%
22%
Proportion of Lodgement Methods for Students Dataset
A
S
Figure 3.1: Pie chart showing proportion of Lodgement Methods for Students Dataset
Table 3.2: Proportion of Tax Agents of Students Dataset
Sample Size 150
Count of Successes 117
Confidence Level 95%
Sample Proportion 0.78
z Value 1.9600
Standard Error of the Proportion 0.033823069
Margin of Error 0.0663
Calculations for Computation of Confidence Interval
Interval Lower Limit 71.37%
Interval Upper Limit 84.63%
To test whether the proportion of Australian people (p1) and International students (p2)
preferring tax agents are equal, z-test has to be computed (Park 2015). The null and the alternate
hypothesis for this test are defined as follows:

5BUS708 STATISTICS AND DATA ANALYSIS
Null Hypothesis (H0): p1 – p2 = 0
Alternate Hypothesis (H1): p1 – p2 ≠ 0
The difference p1 – p2 is denoted by p. From the results of the test given in table 3.3, it can be
seen that the p-value is 0.212 which is greater than the level of significance (0.05). Thus, the null
hypothesis is accepted. Thus, there is no difference in the proportion of Australian people and
International students preferring tax agents for lodgement of tax returns.
Table 3.3: Test for equality of proportion of people appointing Tax Agents from the two datasets
Null Hypothesis H0: p 0 0%
Alternative Hypothesis HA: p <> 0%
Test Type Two
Level of Significance 0.05
Number of Samples for Group 1 1000
Number of Successes for Group 1 732
Number of Samples for Group 2 150
Number of Successes for Group 2 117
Hypothesized Difference 0
Proportion for Group 1 0.732
Proportion for Group 2 0.78
Average Proportion 0.738261
Difference in Two Proportions -0.048
Critical Z value -1.24709
p-value 0.212364
Result of the Analysis
Do not reject Ho
4.0 Age Range and Lodgement Method
The comparison of different types of lodgement methods with respect to different ages is
summarized in table 4.1. The comparison is also shown clearly in figure 4.1. From the analysis, it can be
seen that people for all the age ranges prefer lodging their tax returns with the help of tax agents.
Table 4.1: Comparison of Age Range and Lodgement Method for ATO Dataset
Count of Lodgment_method Column Labels
Row Labels A S Grand Total
0 40 21 61
1 33 6 39
2 52 13 65
3 65 14 79
4 69 17 86
5 92 15 107

6BUS708 STATISTICS AND DATA ANALYSIS
6 82 24 106
7 74 37 111
8 66 35 101
9 72 35 107
10 55 34 89
11 32 17 49
Grand Total 732 268 1000
Figure 4.1: Bar Graph comparing Lodgement method according to Age Range
The association between the two variables age range and lodgement method has to be tested.
This test can be done using the chi square test of association (Gilbert and Prion 2016). The expected
frequencies that are necessary for performing the test is given in table 4.2. The null and the alternate
hypothesis for this test are defined as follows:
Null Hypothesis (H0): There is no existence of significant association between the two variables.
Alternate Hypothesis (H1): There is existence of significant association between the two variables.
From the significance given in table 4.3 which is obtained from the analysis, it can be seen that
the value is less than 0.05. Thus null hypothesis is rejected. There is existence of relationship between
age group and lodgement methods.
Table 4.2: Expected Frequency table for Age Range and Lodgement Method
Expected Frequency
Row Labels A S Grand Total
0 44.652 16 61
1 28.548 10 39
2 47.58 17 65

7BUS708 STATISTICS AND DATA ANALYSIS
3 57.828 21 79
4 62.952 23 86
5 78.324 29 107
6 77.592 28 106
7 81.252 30 111
8 73.932 27 101
9 78.324 29 107
10 65.148 24 89
11 35.868 13 49
Grand Total 732 268 1000
Table 4.3: P-Value for the test of Association
Chi Square Significance Value 0.000156
5.0 Lodgement Method and Total Income
It can be seen from table 5.1 that the average of the total income is higher for the people
appointing tax agents. Figure 5.2 shows the variation in the total income of the individuals. It can be
seen that there is huge variation in the incomes. The incomes of the individuals are not at all close to the
average income. A large number of people are earning more than the average income. There are 55
people preferring tax agents whose income is much higher than the usual income of the people
belonging to that group and 16 people preferring self-preparers whose income is much higher than the
usual income of the people belonging to that group
Table 5.1: Average Income for each type of Lodgement Methods
Row Labels Average of Tot_inc_amt
A 70547.08607
S 46670.23881
Grand Total 64148.091

8BUS708 STATISTICS AND DATA ANALYSIS
Figure 5.1: Bar graph comparing the average income of different lodgement methods
Table 5.2: Summary of total income of different lodgement methods
Measures Tax Agents Self-Preparer
Mean 70547.0861 46670.23881
Standard Error 6665.38537 2624.92117
Median 46564.5 38863.5
Mode 16778 0
Standard Deviation 180335.324 42971.81156
Sample Variance 3.25E+10 1.85E+09
Kurtosis 364.396143 7.003206861
Skewness 17.0865485 2.10474629
Range 4153282 311090
Minimum -6234 0
Maximum 4147048 311090
Sum 51640467 12507624
Count 732 268
First Quartile 23185.75 16859.5
Third Quartile 75541.5 62481.75
Interquartile Range 52355.75 45622.25
Calculation of Outlier Range
Lower Outiler Range -29170 -28762.75
Upper Outlier Range 127897.25 108104
Number of Outliers
Number of Outliers 55 16

9BUS708 STATISTICS AND DATA ANALYSIS
Figure 5.2: Boxplot showing shape of the distribution
6.0 Total Income and Total Deduction based on Lodgement Methods
Positive relationship exists between total income and total deduction of the people preferring
tax agents but it can be seen from the r square value given in table 6.1 that the relationship is very weak.
Only 2 percent of the variations in deductions can be explained by income. The relationship can be
expressed with the help of the following equation:
Deduction = (0.0057 * Income) + 2431.8
Figure 6.1: Relationship between Income and Deduction for Tax Agents
Table 6.1: Regression Statistics (Tax Agents)

10BUS708 STATISTICS AND DATA ANALYSIS
Multiple R 0.13
R Square 0.02
Adjusted R Square 0.02
Standard Error 7661.26
Observations 732
Table 6.2: ANOVA (Tax Agents)
df SS MS F
Significance
F
Regression 1
7.76E+0
8 7.76E+08
13.22
9 0.000
Residual 730
4.28E+1
0
5869494
5
Total 731
4.36E+1
0
Table 6.3: Regression Coefficients (Tax Agents)
Coefficient
s
Standard
Error t Stat P-value Lower 95% Upper 95%
Intercept 2431.813 304.093 7.997 0.000 1834.813 3028.814
Tot_inc_amt 0.006 0.002 3.637 0.000 0.003 0.009
Positive relationship exists between total income and total deduction of the people preferring
themselves for lodgement of tax returns but it can be seen from the r square value given in table 6.4
that the relationship is moderate. Only 26 percent of the variations in deductions can be explained by
income. The relationship can be expressed with the help of the following equation:
Deduction = (0.049 * Income) + 779.63

11BUS708 STATISTICS AND DATA ANALYSIS
Figure 6.2: Relationship between Income and Deduction for Self-Preparers
Table 6.4: Regression Statistics (Self-Preparers)
Multiple R 0.51
R Square 0.26
Adjusted R Square 0.26
Standard Error 3543.75
Observations 268
Table 6.5: ANOVA (Self-Preparers)
df SS MS F
Significance
F
Regression 1
1.18E+0
9 1.18E+09
94.21
1 0.000
Residual 266
3.34E+0
9
1255816
2
Total 267
4.52E+0
9
Table 6.6: Regression Coefficients (Self-Preparers)
Coefficients
Standard
Error t Stat P-value Lower 95% Upper 95%
Intercept -779.628 319.903 -2.437 0.015 -1409.491 -149.765
Tot_inc_amt 0.049 0.005 9.706 0.000 0.039 0.059
7.0 Conclusion
From all the analysis conducted in the above sections, it can be concluded that both the people
of Australia and the International students has a preference for appointing tax agents for the lodgement
of tax returns. There is no difference in the proportion of the Australian people and the International
students preferring tax agents. Relationship has been found between age range and preference of
lodgement methods. The people who are tax payable earn more than the average income. There is
positive relationship between income and deduction amounts for both the types of lodgement methods
but the relationship is weak.
The effect of gender on the lodgement methods have not been analyzed so far in this research.
This can be conducted as further research.

12BUS708 STATISTICS AND DATA ANALYSIS
References
Bitler, M., Hoynes, H. and Kuka, E., 2016. Do In-Work Tax Credits Serve as a Safety Net?. Journal of
Human Resources.
Gallemore, J. and Labro, E., 2015. The importance of the internal information environment for tax
avoidance. Journal of Accounting and Economics, 60(1), pp.149-167.
Gilbert, G.E. and Prion, S., 2016. Making Sense of Methods and Measurement: The Chi-Square
Test. Clinical Simulation in Nursing, 12(5), pp.145-146.
Park, H.M., 2015. Hypothesis testing and statistical power of a test.