Coursework Maths - Mayfield HighWatch
i'm new here and i'm in yr 10!!!
i'm stuck on my yr10 maths coursework ... which is a statistical coursework on a school called mayfield high (made up.. based on real data!) ... i need help from the start to the end ... i'm doing higher maths as well ... but dont understand what i need to do ... because of my teacher who doesn't explain things good enough
i understand that i need to choose hypothesis and then sample size and then the sample itself..... i would appreciate some help or tip on the start... which seems important.
i'm stuck on testing the sample for normality right now... please help.. and God knows what when i tackle this
i'll be grateful.. thanks
I would advise the following procedure, I have added notes too.
It looks long but I have broken it down a lot.
Make an hypothesis:
I recommend suggesing that girls hit puberty before boys, and that therfore the weights, and heights of Y7 girls are more than the boys, but by Y11 the positions have stopped as more boys hit puberty.
This is good for higher as it allows you to compare the two different variables as well as the difference between boys and girls.
Take a random sample:
Use the random button on your calculator, or roll a dice to chose random records in the database from Y7 and Y11, you will use these to draw up your first graphs.
In reality you can just pick them yourself, but pretend the above.
Make sure that you describe the procedure that you take thoroughly.
First set of graphs:
Here you should draw one for the boys' weight against height and a similar one for the girls.
Stick to a standard scatter graph and keep the same scale for both.
Add a line of best fit to both.
Now you can compare the contents of the two graphs.
You can work out the gradients of the lines of best fit and point out there significance.
Make sure you compare how it looks on the graph, as well as what this means in real life.
Eg. Girls has a greater spread of data, this shows that there is greater variatioon at this age in height and weight at this age between girls than boys.
This may seem like repeating yourself, but it is what they want.
If you know what these are you may want to draw two to show the data in a different way, they show the range and interquartile range quite nicely.
Conclude on first graphs:
Say what you have learnt from your first graphs.
Do you hypothesises seem to be correct?
Then you will probably need to state that this requires further investigation in the form of a stratified sample.
Then you do that sample.
This is a sample where the amounts of different groups pof people in the whole group are reflected with the number of people chosen from each section.
I will continue this guide if it helps, please respond and let me know.
i understand the things u've written (boxplot etc..) ... i need to go in order, and testing my sample for normality at the moment is my biggest problem... i basically don't know what things (in order) i need to do to see if my sample is normal... for example do i put my data in frequency tables draw cumulative graphs, find median and draw boxplots.. and if the result is ok i can continue with other tests, and if not i need to choose another sample ?;
what i'm trying to say, what tests do i do (in order) for normality, if someone could tell me like below for example:
1st - draw cumulative graph, find median and draw box plot
2nd - draw histogram find whatever
3rd ... and etc.... all just to make sure the sample is normal
do u understand me ?
how many tests do i need to do, what do i need to find out that shows my sample is good .. and how, by drawing histograms, boxplots ?? what ???...
i would appreciate it if someone could help me step by step .. not in the actual calculations and conclusions (well not at the moment at least).. but in the whole plan... an accurate plan which i could look at and say right i need to draw 'whatever' and find 'whatever' , then i need to find 'whatever' by drawing 'whatever'... and so on
hope i haven't confused anyone...
sam... thanks and if u could help me more i'll be grateful