Factors affecting price of Used Car

Authors Avatar

Introduction

I have been given a coursework with secondary data that contains different types of car makes, each with different models. The data has all the mandatory information of a car. These are; colour, age, mileage, engine size, both price when new and price when used. My objective is to find out which of the factor/s mentioned above has/have the most effect on the price of a used car.

Hypothesis

A car is considered to be used when it has been travelled with. Travelling consumes time and consequently increases the mileage of a car. I therefore predict that the older a car gets, the more distance it covers and hence undoubtedly the less it will cost in the market.


Method of Enquiry

I am going to carry out an investigation applying all my mathematical and general knowledge to analyze and find out the factor/s affecting the price of a second hand car. Since the data I have is of two types i.e. Qualitative data (involves names and words e.g. black, blue, yes and no) and Quantitative data (involves numbers and figures), this makes it very complex to work on. I will refine my data so that I will work on only the major factors, which affect most and then later on I will break them further into small bits so that I will remain with a maximum of two factors. The reason for doing this is because some of the information contained in the data is hardly of any use to help me complete my objective e.g. air condition and gearbox. With factors like these, the range difference between two different makes is very small such that it has barely any effect on the price of a use car.


Procedure

I will start my investigation with the qualitative data, these are colour and car make. For each of them I will find the most popular colour and car make. I will record my results in a tally table and a frequency table so that I can easily spot the most popular ones. I will then use the 4 most popular car makes from the tally chart of all of the cars as my independent variable this is because the thing that people look at first when they are buying a car is the make and I believe this will affect the second hand price the most. Certain cars have a very good reputation of being reliable over others. Also some cars have a higher social status than others for example people would prefer a Porsche to a Ford.

However the 4 most popular car makes add up to almost half the total number of cars, I therefore need to reduce this number to get reliable and valid results. I will need to refine it so that I get a sample that would be representing a portion of the cars. I will carry our stratified sampling. The reason I am doing this is because working on about 50 cars will be difficult to analyse and also the results will be unreliable and invalid. From there I will be looking at the averages of the second hand price of each car make. The reason I will be doing this is to find the measure of spread. I will then precede on to plot a graph of each factor against each car make of which I will compare them with each other see if there is any relationship.

        I am then going to find the standard deviation because it measure spread of the data about the mean value. The reason for doing this is because it will give me a more detailed picture of the way in which the data dispersed about the mean as the centre of distribution. Its main use is to compare two sets of data. If a set has low standard deviation, the values are not spread out too much.

        My final stage will be further analysis of the correlation between each of the three factors I chose and percentage difference between price when new and second hand price. I will look at each car makes coefficient of correlation represented by the symbol ‘r’ which will tell me exactly what the correlation is using figures. If the correlation is 1 or -1 then the line of best fit is on all the plotted points, however if ‘r’ is 0 then none of the points is on the line of best fit

Most Popular Car

Since the data I have is not in a specific order, I will have to organise and sort them in a descending order by colour. It will be much easier counting using this method. The results I got are recorded on a Tally and Frequency table as shown below.

From the table above, I can notice that red seems to be the most popular colour out of the 100 cars with a frequency of 23, while the least popular colours are Aubergine, Burgundy, Cuirass, Gold, Marine, Night fire, Purple, Tourmaline with a frequency of one.

The following charts represent the type of colour and their frequency.

A Column chart showing Colour against Frequency            

A Pie chart showing Colour against Frequency 

The Most Popular Car Make

I am now going to find the most popular car make. I will have to organise and sort them in a descending order by car make. The results I got are recorded on a Tally and Frequency table as shown below.

From the results on the table, I can clearly see the most popular car make which is Ford because it has a frequency of 16 out of 100 while the least popular car makes are

Audi, Bentley, Honda, Landover, Lexus, Mazda, Porsche, Rolls, Royce, Seat, Suzuki, Toyota, all with a frequency of one.

I am going to represent these results in charts to show the popularity of cars.

A Column Chart showing Car Make against Frequency

This above graph also shows that Ford is the most popular car followed by Vauxhall

From chart it shows that there are four clear favourites for the most popular car and that Ford is the most popular, followed by Vauxhall, Rover and Fiat.

Choosing the appropriate data to work on.

As I said previously on my planning that I will choose car make to carry out my investigation rather than using the most colour. This is because car make is mainly the first priority of a person due to its engine size, shape and other factors but he/she can easily go and paint the car with the colour he desires and it wouldn’t affect the car make.

Considering the fact that the sample size I have is quite small, I will choose the top 4 cars makes with most frequency. These are Ford, Vauxhall, Rover and Fiat respectively. I believe these 4 car makes add up to 51 cars; 16+13+12+10=51 which is more than half the total number of cars. I will then carry out stratified sampling. I will select stratified random sample of size 40. By using stratified sampling I will ensure that my sample is representative of the 100 cars.

Since the sample size I need is 40 cars. I will calculate what proportion of each of the 4 car makes represents the sample size. This is how I will do it.

Join now!


Therefore the proportions for each car make will be: No. of Cars for each Car Make     X 40

                                                          Total No. Of Cars for all Car Make

 

  .         Ford:                 16/51 X 40 = 13

.   .        Vauxhall:        13/51 X 40 = 10

        Rover:                12/51 X 40 = 9

        Fiat:                10/51 X 40 = 8

Now that I know how many cars of each car make I need; I will make a random selection. The reason for doing this is to avoid any biasness.

I will number each model for all ...

This is a preview of the whole essay