This project is about investigating the data about an imaginary school. Data for all students is given and these data includes eye colour, weight, Exam Results, the Average hours of TV per week, Height, Results in KS2 and many more. Ozioma Ihecherenoma SPECIFY THE PROBLEM and PLAN HYPOTHESIS I have decided to find out the following: 1. If the IQ of a student affects their KS2 Total: I think that the higher the IQ, the higher the KS2 Total because people with a higher IQ tend to remember things better than those who have a low IQ. 2.
If the Average hours of TV watched per week affects their KS2 Total: I think that if people spend too much time watching TV, they will not study and will end up having a low KS2 Total. 3. Comparing the boys IQ and their KS2 Total against that of the girls IQ and their KS2 Total: I think that boys who have a high IQ will have a high KS2 Total and it will be greater than the girls will. The reason that I have chosen to investigate these factors is that I feel that the IQ and Average hours of TV watch per week affects the result of any student who decides to take an exam.
I expect my research to show me a positive correlation between IQ and KS2 results and a strong-positive correlation between Average hours of TV and KS2 Total. The reason that I believe this is because high IQ’s yields higher results and that if you watch too much TV; you will bring your result down. SAMPLING To complete this task, I will need to collect data that has things to do with IQ, Average TV watched and KS2 Total. The number of data that I will collect and use for this investigation is 80 pieces. I will use all of the data that I collect.
The reason why I have decided to use my entire sample is that I feel it will give me a representation of the proportion of the year group that I am investigating. To get my sample, I will use a stratified random sample. This means that I will use a proportional number to represent each year group (e. g. 40) and then pick students at random. This proportional number is the total number of samples that I want to use in this investigation. Next, I will find the fraction of students in the year group and times it by 40. For example, I will take the total number of year 7s and divide it by the total number of students in the school.
It was then times by 40. Each year group had samples gotten by using this process. The reason why I have chosen this method is that I feel this is a good way to get a reliable proportion, which will represent my research. I will record my results by using methods such as calculations, tables, graphs (scatter graph), Spearman’s Rank Correlation, print screen shots and lastly, explaining the workings that I have used during the investigation. The types of graphs that I will draw for this investigation are Scatter Diagrams, Spearman’s Rank Correlation and Standard Deviation.
COLLECT, PROCESS, and REPRESENT Below are my stratified samples. These are the number of students that will represent each year group. However, I need to find out the proportion of males to females that I will use in each year group. This is because I will need a good proportion to represent the males and the females in each year group. Below are the proportion for males and females in each year group. This is the formula that I will use to work out Spearman’s Rank Correlation, where n= number of sample. This graph is comparing the KS2 Total against the IQ.
From the graph, I can see that there is a strong positive correlation. This means that my first hypothesis is certainly right because a high IQ yields a high KS2 Total. However, to be very confident with my hypothesis, I will carry out a Spearman’s Rank Correlation Test to check whether these two factors actually correspond with each other. I have drawn a line of best fit and this gives me an accurate prediction about the correlation of the comparisons that I am carrying out. This is Spearman’s Rank Number. Since my data was too much, I print-screened the important part, which was the actual calculations itself.
When I carried out Spearman’s Rank Correlation, I got a strong-positive correlation, which meant that IQ has an effect on KS2 Total. This also tells me that my scatter graph is right because I got a strong-positive correlation and this is what I got when I carried out my Spearman’s Rank Correlation Test. So therefore, now it is very clear to me that if someone has a high IQ, the student is very likely to get a high KS2 Total. I have created this graph because it can help me to predict data that is not on the graph. Instead of using a line of best fit, I used an Exponential line, which is a curve that helps predict data.
Therefore, from using this exponential line, I can predict that a student with a KS2 Total of 16 could have an IQ of about 125, while a student with a KS2 Total of 20 could have an IQ of 130. This graph is comparing the KS2 Total against the Average hours of TV watched. From the graph, I can see that there is no correlation. This means that my second hypothesis is certainly wrong because the graph tells me that it does not matter the amount of hours spent in front of the television because they can still get a high KS2 Total.
However, to be very confident with my hypothesis, I will have to carry out a Spearman’s Rank Correlation Test to check whether these two factors actually work with each other. I have drawn a line of best fit and this gives me a rather accurate correlation of the comparisons that I am carrying out. This is Spearman’s Rank Number. Since my data was too much, I print-screened the important part, which was the actual calculations itself. When I carried out Spearman’s Rank Correlation, I got no correlation, which meant that the Average hours of TV watched in a week has no effect on KS2 Total.
This also tells me that I interpreted the scatter graph rightly because I thought that there is no correlation. So therefore, now it is very clear to me that it does not matter if someone spends too much time watching the telly because the student can still have a high KS2 Total. Overall, I would say that this calculation supports my hypothesis. The line used here is a Logarithmic line. This line also helps predict data, just like the exponential line; but the difference is that the exponential line goes up while the logarithmic line goes down.