Statistics and Data Analysis for Fuel Prices in Australia

Statistics
Statistics and data analysis
Student name:
Tutor name:
1 | P a g e

Statistics
Executive summary
The objective of this research was to compare the fuel prices among different brands in
Australia. A list of 1000 stations from different suburbs was involved in the sample and the price
of petrol in brands in those suburbs recorded. Various variables were involved but the most
important ones were the brand and the price of petrol. The other variables were the suburb that
represents the town where the brand is located while the brand represents the company that is
selling the fuel. Descriptive statistics was used to describe the data. Measures of central tendency
such as the mean, median and modal petrol price were calculated and found to be 123.03, 122.15
and 119.9 respectively. Measures of dispersion such as variance and standard deviation for the
petrol price were also calculated and found to be 173.39 and 13.16 respectively. The minimum
and the maximum prices were 56.9 and 185.6 respectively. The research study tested the
hypothesis that the petrol prices in service stations in September 2016 were less than 115. The
research found that there was no sufficient evidence to support the claim hence it was concluded
that the average price of petrol in all service stations in September 2016 is less than or equal to
115 Australian cents. The research also had another dataset showing the preferred stations for the
students. A sample of 30 students from KOI was sampled using simple random sampling and
asked about their preferred petrol stations. 10 students representing 33.3% preferred buying fuel
from Caltex. 7 students preferred buying the fuel from Caltex Woolworths representing 23.3%. 4
students representing 13.3% preferred buying from 7-Eleven. 3 students preferred BP and Coles
Express representing 10% each. 2 students preferred United representing 6.7%. 1 student
preferred Speedway representing 3.3%.
2 | P a g e

Statistics
Introduction
Fuel prices have been a subject of discussion in the recent years. The prices have been staggering
due to the price of crude oil which is dictated by the oil producing and exporting countries
(Giradi 2012) and (Elliot 2015). This is coupled again by the rate at which the dollar is trading in
these particular countries. In various local countries, the cost of the final product which is fuel is
further escalated due to various factors such as cost of importation and proximity to the port of
entry (Williams 2011). Towns and cities which are near the pipeline buy the fuel at a little lower
rate per litter than those that are far away from the pipelines (Kraus 2017). Suburbs in Australia
have also been facing fluctuation of fuel prices due to the factors mentioned above. For that
reason different towns and brands have been selling fuel at different costs per litter.
According to weekly global fuel prices review (May 7, 2019) (www.globalpetrolprices.com)
there was a two week of decline of crude oil around the mentioned. In contrast, the review
reported an increase in the price of gasoline. The international crude oil prices recorded that the
global mean gasoline and diesel costs were 1.13 and 1.04 US dollars in every liter respectively.
The report also showed that the European average gasoline cost was 1.43 US dollars recording a
0.5% increase in the same compared to the previous week. In Asia, the price also went up by
1.5%. In Africa and USA, it went up by 0.6% and 0.2% respectively. It was also found that the
major increases were experienced in New Zealand, Thailand and Croatia (Macrotrends 2017).
Dataset 1
Data set 1 contains six variables. Three of them are nominal variables which include address,
suburb and brand. Address in this case represents the directions of the location where a particular
brand is located. Suburb represents the town where the brand is located while the brand
represents the company that is selling the fuel. The other variables are ordinal variables which
include postcode and fuel code. The only numerical variable which is very important for analysis
is the fuel price. Price in this case is a numerical variable which is also an interval measure. The
data is a secondary type of data since it is obtained from a source and not collected first hand.
3 | P a g e

Statistics
Dataset 2
Dataset 2 about KOI students’ preference of service station. In selecting of the students from a
list, simple random sampling and convenience sampling were used to pick the students that have
made the sample. Convenience sampling was used to select 90 students from the population who
admitted that they have been buying petrol. After this, simple random sampling was used to
select the students that finally made the sample. This method (simple random sampling) was
found suitable as it gave each country an equal opportunity of being selected into the sample.
The limitation that this method had is that there are chances that we could have collected samples
from not many variations. The variable “percentage change” involved here is numeric. Since
data was extracted from a review report, it can be concluded that the data is secondary.
Dataset 1 variable analysis
Descriptive statistics for fuel pricePrice
Mean 123.0396
Standard
Error
0.416409
064
Median 122.15
Mode 119.9
Standard
Deviation
13.16801
08
Sample
Variance
173.3965
083
Kurtosis 1.836927
633
Skewness -
0.148005
63
Range 128.7
Minimum 56.9
Maximum 185.6
Sum 123039.6
Count 1000
Table 1
4 | P a g e