GG2038 Assignment 3: Investigating UCC Student Commuting Patterns

This report presents an analysis of the commuting patterns of University College Cork (UCC) students, focusing on greenhouse gas emissions and climate change implications. The study utilizes descriptive and inferential statistical methods to examine student commuting distances, modes of transport, and related factors. Descriptive statistics, including bar charts and pie charts, are used to analyze commuting distances for different student groups (e.g., first-year, third-year, BA, BSc students) and their sensitivity to carbon footprint. Inferential statistics, such as t-tests and chi-square tests, are employed to determine significant differences in commuting distances between student groups and to investigate the relationship between commuting mode and student level. The analysis includes hypothesis testing, critical value determination, and decision-making based on statistical outputs. The report also discusses the methodology, results, and potential improvements for future research, including the use of stratified sampling and the provision of residential facilities to reduce carbon footprint. The study concludes with recommendations for UCC to promote sustainable commuting practices, such as encouraging cycling, providing bus services, and fostering carpooling to minimize individual car travel and its environmental impact.

QUANTITATIVE GEOGRAPHY
STUDENT ID:
[Pick the date]

Part 2: Descriptive Statistics
The respective descriptive statistics related to the commutation distance for the variable
student needs to be computed. For the students as a whole, the descriptive statistics related to
commutation distance are pasted below.
With regards to the first year undergraduate students, the descriptive statistics related to
commutation distance are pasted below.
With regards to the third year undergraduate students, the descriptive statistics related to
commutation distance are pasted below.
The bar chart with regards to the sensitivity of the UCC students in relation to carbon
footprint is summarised in the following graph.

From the above responses, it is apparent that the students at UCC do not tend to very
sensitive with regards to carbon footprint with regards to commutation. In this regards,
maximum responses comprise of those individuals who do not consider the carbon footprint.
However, there are sizable number of students who tend to be quite sensitive about carbon
environment while making the commutation choice. Also, there are students who sometimes
tend to consider carbon footprint while making the commutation choices to the university.
The requisite pie charts are illustrated as follows.

Based on the above pie-charts, it is apparent that a large proportion of students residing in
rented accommodations tend to walk to the university. One of the possible reasons for the
same is that these students would be renting near the university campus only so as to
minimise the transportation cost and time involved. In sharp contrast, the mode of
commutation for those who live with family is significantly more diverse with walking
percentage a lot lower. This is not surprising as the respective houses of most students would
not be in close vicinity to the university and hence walking would not be a viable choice.
The requisite histogram is as highlighted below.

There does not seem to be any significant difference between the commutation distances
between the students who participate or do not participate in UCC Clubs and Societies. The
apparent difference between the two is on account of scale differences which once adjusted
would result in similar distribution. Additionally, neither of the above distributions is normal
considering the fact that right skew is present for both the distributions indicated above
(Medhi, 2016).
PART 3: Inferential Statistics
The requisite steps in the given hypothesis testing are performed below.
Step 1: Defining the null hypothesis
H0: μBA=μBSc i.e. there is no significant difference in the average commuting distances
between the B.A. students and B.Sc. students.
Step 2: State the alternative hypothesis
Ha: μBA≠μBSc i.e. there is significant difference in the average commuting distances between
the B.A. students and B.Sc. students.
Step 3: The significance level for this test has been assumed as 5% which would be
appropriate for the given hypothesis where higher accuracy would not be required.
Step 4: The relevant test statistic in the given case would be t statistics. This is because the
given for the two samples i.e. commuting distance of B.A. students and commuting distance
of B.Sc. students, the population standard deviation is not known. If the population standard
deviation was known, then the relevant test statistics would have been z (Flick, 2015).
Considering that there are two samples and both are independent of each other, hence a two
sample independent t test would be conducted. Under this category, there are two options in
the form of equal and unequal variance. Considering that the sample size of the two samples
is not equal, hence unequal variance is assumed (Hair et. al., 2015). The value of the test
statistic can be obtained from the excel output indicated as follows.

It is apparent from the above output that the t statistics has come out as 0.344.
Step 5: The critical values for the test need to be highlighted. The given test is a two tail test
as the alternative hypothesis contains the “not equal to” sign. As a result, there would be two
critical values, one at the lower end and the other at the upper end. From the above output,
the t critical value at the upper end is 1.9655 and the corresponding value at the lower end
would be -1.9655 (Hillier, 2016).
Step 6: The decision rule with regards to use of critical value is that if the test statistic lies
within the interval defined by the critical values, then the null hypothesis would not be
rejected and alternative hypothesis would not be accepted. In the given case, the computed t
statistic is 0.344 which tends to lie between -1.9655 and 1.9655. As a result, the available
evidence does not warrant null hypothesis rejection (Lieberman et. al., 2013). Hence, it can
be concluded that there is no significant difference between the average commutating
distance of B.A. and BSc students.
Part 4: Extra Analysis
The research question that has been chosen for this analysis is as follows.
“Is there any relationship between the mode of transport used for commutating to the
university and the level of students?”
The rationale for choosing the above research question was to critically analyse if the level of
students tends to influence their choice of transport system. Thereby, the research question
aims to explore the preferences of the two levels of students in the context of their
commutation choices which then can be further analysed to explore the underlying reason
along with the role of potentially other variables such as distance of commutation along with
location.

The relevant statistical analysis would constitute of an inferential test as the underlying
objective is to highlight the population parameter based on the given sample statistics.
Hypothesis testing would be deployed and the appropriate test in this regards would be Chi-
square test of independence. This test is appropriate as both the variables of interest have
categorical measurement of the data and the underlying data type is non-numerical. Under the
given situation, the chi square test statistics would be the most appropriate (Hastie, Tibshirani
and Friedman, 2014).
The requisite hypotheses for this test are indicated below.
Null Hypothesis: There is no significant relationship between the commutation transport
mode and level of course.
Alternative Hypothesis: There is significant relationship between the commutation transport
mode and level of course.
The level of significance for the given hypothesis test has been assumed as 5%.
In order to compute the chi-square statistic, the first step is to indicate the actual frequency of
the actual mode of transport divided in accordance with the level of student.
The next step is to use the above data for obtaining the expected frequencies of the usage of
the various modes of commutation in accordance with the level. This is indicated below.

Comparing the actual frequencies and expected frequencies, we can obtain the value of the
test statistic i.e. chi-square which has been highlighted below.
The p value has been obtained considering the chi square test statistic (21.05) from the above
computation along with the degrees of freedom (9-1)*(2-1) = 8. This p value has come out as
0.007 using Excel as an enabling tool.
Comparing the p value obtained above with the level of significance (0.05), it is apparent that
the available evidence is sufficient for causing rejection of the null hypothesis. As a result,
the alternative hypothesis would be accepted (Hillier, 2016). The acceptance of alternative
hypothesis indicates that the commutation mode and the level of students are inter-related.

The inter-relationship between the student level and the commutation means to the university
ness to be further analysed considered the impact of distance from university which might be
impacting the relationship that is observed in the given hypothesis testing. Also, any
empirical support from existing literature needs to be explored with regards to offering
explanation in this regards which would lend more clarity on the given relationship (Eriksson,
and Kovalainen, 2015).
Part 5: Discussion
The objective of this section is to critically review the methodology and the underlying
results obtained.
1) The questionnaire for the given survey worked real well. One of the key reasons for the
same was that the questions were framed in an objective manner and the suitable options
were presented to the respondents to choose from. This allowed the respondents to devote a
lesser time to fill the survey and ensured that the responses were accurate owing to lack of
underlying subjectivity in questions which is typical in open ended questions. With regards to
improvements, it was noteworthy that very few comments were collects in any other
comments. In the survey, mechanisms need to be incorporated so as to enhance the responses
in this regards as it provides vital information that could be useful.
Another potential improvement is the questionnaire is to reduce the number of questions
especially if there are questions which have limited relevance as the respondents highlighted
concerns with regards to the length. Further, it would have been helpful if the respondents
had been briefed before the survey with regards to the purpose of survey and also the various
questions. This would potentially ensure in enhancing the accuracy of the data collected from
the respondents. Besides, it would be a good idea to pass on the survey to the respondents and
allow them some time to fill the same. This is especially critical in the context of certain
information particularly the distance of commutation to the UCC which not everyone would
be aware of with accuracy.
Another crucial aspect is the sample which has been used for acting as the respondent for the
survey. This is critical considering the fact that if the sample is not representative of the
population (i.e. students at UCC), then the reliability of the underlying results would be
adversely impacted. Attempt had been made for the current survey to ensure that all the key
attributes of the population were included in the given list of respondents which was

randomly selected. However, going forward, it would make more sense to deploy stratified
sampling which would result in a more representative sample and hence enhance the
reliability of the trends observed.
2) One of the interesting aspects in relation to commutation trends is that for rented students,
more than 75% tend to walk for commutating to the university. Hence, the key factor which
tends to determine the choice of commutation mode is the distance from the university.
Therefore, in order to reduce the carbon footprint, it is imperative that the UCC must invest in
providing some hostel or residential facility either on the campus or near the campus. This
should be especially lucrative to outstation students and also for those whose home is quite
far from the campus.
Additionally, it may make sense for the UCC to encourage the students to use bicycles as a
mode of commutation. This can be done by providing parking for the same inside the campus
and not allowing any other vehicle inside the campus. Further, the university may also
deliberating on plying buses if certain routes can be identified where maximum students can
have assess and this would ensure that maximum students can travel together in the same
mode of conveyance. The important aspect is to discourage individual car travel and hence
car pooling forums should be available on the UCC student discussion so that the students
can coordinate amongst themselves to reduce their transportation cost and also the carbon
footprint in the process.
Out of the various choices suggested above, the most effective is likely to be one that
involves providing residential facilities on or near the campus as it would be the most
convenient and also ensure that the students would be able to save money.

References
Eriksson, P. and Kovalainen, A. (2015) Quantitative methods in business research. 3rd ed.
London: Sage Publications.
Fehr, F. H. and Grossman, G. (2013). An introduction to sets, probability and hypothesis
testing. 3rd ed. Ohio: Heath.
Flick, U. (2015) Introducing research methodology: A beginner's guide to doing a research
project. 4th ed. New York: Sage Publications.
Hair, J. F., Wolfinbarger, M., Money, A. H., Samouel, P., and Page, M. J. (2015) Essentials
of business research methods. 2nd ed. New York: Routledge.
Hastie, T., Tibshirani, R. and Friedman, J. (2014) The Elements of Statistical Learning. 4th
ed. New York: Springer Publications.
Hillier, F. (2016) Introduction to Operations Research. 6th ed. New York: McGraw Hill
Publications.
Lieberman, F. J., Nag, B., Hiller, F.S. and Basu, P. (2013) Introduction To Operations
Research. 5th ed. New Delhi: Tata McGraw Hill Publishers.
Medhi, J. (2016) Statistical Methods: An Introductory Text. 4th ed. Sydney: New Age
International.