GEOG2132: WHM Spending in Australia Based on Age & Education Level

This report investigates the research question of whether the spending level of Working Holiday Makers (WHMs) in Australia is influenced by their characteristics, specifically age and education. The analysis, based on a dataset of 19,888 individuals, reveals that 54.3% hold a university degree, and the majority fall within the 20-28 age range. Spending levels are categorized into five quantiles, with most individuals concentrated in the third and fourth levels. Cross-tabulation analysis indicates that while education level doesn't show a clear pattern with spending, there is a gradual increase in spending with age. Statistical testing confirms a significant positive association between both age and education (recoded into high/low categories) with spending habits, suggesting that older, more educated individuals tend to spend more. The report concludes that age and education are indeed influential factors in determining the spending patterns of WHMs in Australia.

Running head: A1719342
GEOG2132 &1005: Social Science Techniques
Name of Student
Name of University

1A1719342
Table of Contents
Question 1.............................................................................................................................................3
Question 2.............................................................................................................................................5
Question 3.............................................................................................................................................8
Appendices..........................................................................................................................................11

2A1719342
Question 1
The assignment addresses the research question that whether spending level of
WHMs in Australia influenced by the characteristics of respondents, for example, age, sex,
citizenship, marital status, educational level. The two characteristics chosen for this paper is
age and education. Therefore focussing on these two variables the situation of age and
education being taken into account in the dataset is scrutinized. The following table describes
the level of education, the age and the spending habits of the individuals who constitute the
sample of data. No missing values were found in the sample of data.
Frequencies
Statistics
What is your
highest level of
education?
5 quantiles of
spending Reduced Age
N Valid 19888 19888 19888
Missing 0 0 0
54.3% of the individuals in the sample have university degree and , 8.6% has a trade
qualification, 12.6% have some other form of post school qualification, 22.6% had completed
high school and 1.9% were found to have not graduated high school.
Frequency Table
What is your highest level of education?
Frequency Percent Valid Percent
Cumulative
Percent
Valid University degree 10800 54.3 54.3 54.3
Trade qualification 1711 8.6 8.6 62.9

3A1719342
Other post-school
qualification
2512 12.6 12.6 75.5
Completed high school 4488 22.6 22.6 98.1
Did not complete high
school
377 1.9 1.9 100.0
Total 19888 100.0 100.0
The quantiles of spending captured in the data was found to have 16.8% in the first
level, 16.1% in the second level, 35.1% in the third level. 21.3% in the fourth level and
10.8% in the highest or fifth level. Thus 56.4% of the individuals are in the third and forth
level and about 33% are below that.
5 quantiles of spending
Frequency Percent Valid Percent
Cumulative
Percent
Valid First level (low) 3345 16.8 16.8 16.8
Second level 3194 16.1 16.1 32.9
Third level 6983 35.1 35.1 68.0
Forth level 4228 21.3 21.3 89.2
Fifth level (high) 2138 10.8 10.8 100.0
Total 19888 100.0 100.0
74% of the individuals in the sample are between 20 to 28 years, thus they’re new
adults and only 14.1% are above that age group where as 11.8% are below.
Reduced Age
Frequency Percent Valid Percent
Cumulative
Percent
Valid [16, 20) 2354 11.8 11.8 11.8
[20, 24) 7347 36.9 36.9 48.8
[24, 28) 7374 37.1 37.1 85.9
[28, 33) 2813 14.1 14.1 100.0
Total 19888 100.0 100.0

4A1719342
Question 2
The age of the individuals and the level of education of the individuals was compared
with their level of spending using a cross tabulation table. None of the data observations were
missing.
Crosstabs
Case Processing Summary
Cases
Valid Missing Total
N Percent N Percent N Percent
5 quantiles of
spending * What is
your highest level of
education?
19888 100.0% 0 0.0% 19888 100.0%
5 quantiles of
spending * Reduced
Age
19888 100.0% 0 0.0% 19888 100.0%
It was seen that 69.4% of individuals with university degree fall in at least the third
level of spending. 71.3% of individuals with trade qualification have their spending in third
level at least. 68.8% of those with other post school education have their spending at least in
third level, 59.9% of people with only high school degree and 64.7% with not even a high
school degree have at least spending in third level. Thus no clear pattern can be seen between
education level and spending pattern.

5A1719342
50.1% of those in age group [16,20) were seen to have spend at least at third level,64% of
those in age group [20,24) were seen to have spend at least at third level,73% of those in age
group [24,28) were seen to have spend at least at third level and 74.3% of those in age group
[28,33) were seen to have spend at least at third level. Thus a gradual increase in observed
with age for those who spend at least at 3rd level.

6A1719342
The following table shows how pattern of spending varies over the four age groups. It
is seen t hat for the lower age groups their spending is mostly at the 1st level, and the others
are more tended to be at third level of spending. The most spending at the highest level is
seen to be done by those in [28, 33).

7A1719342
Question 3
Finally, testing for the association between age and spending, the following table
which gives the results, reveal that there is indeed significant positive association between
age and spending.
Chi-Square Tests
Value df
Asymptotic
Significance
(2-sided)
Pearson Chi-Square 742.307a 12 .000
Likelihood Ratio 759.289 12 .000
Linear-by-Linear
Association
590.542 1 .000
N of Valid Cases 19888
a. 0 cells (0.0%) have expected count less than 5. The
minimum expected count is 253.06.
In order to test for the association between education and spending the education
variable was re-coded into having just two categories, high as low. This was done since no
clear pattern could be identified from exploratory analysis using the original five categories.
Then the following figure shows how the spending levels vary as per high or low education
and it is seen that more people with high education spend beyond third level of spending
than those with low level of education.

8A1719342
Using the recoded variable for education and spending, Pearson’s Chi-squared test
showed that there is indeed positive association between the two. The following table shows
the results of the same.
Chi-Square Tests
Value df
Asymptotic
Significance
(2-sided)
Pearson Chi-Square 123.161a 4 .000
Likelihood Ratio 123.802 4 .000
Linear-by-Linear
Association
111.147 1 .000
N of Valid Cases 19888
a. 0 cells (0.0%) have expected count less than 5. The
minimum expected count is 793.04.

9A1719342
Hence it is concluded that the variables and education both are associated with the
spending habits of an individual. An individual who is highly educated and older is thus
expected to spend more than someone who is younger or who is not highly educated or both.

10A1719342
Appendices
5 quantiles of spending * 2 levels education Cross-tabulation
Count
2 levels education
Totallow level high level
5 quantiles of
spending
First level
(low)
1404 1941 3345
Second level 1334 1860 3194
Third level 2517 4466 6983
Forth level 1485 2743 4228
Fifth level
(high)
637 1501 2138
Total 7377 12511 19888