The Student Room Group

Scroll to see replies

I got full marks on the stats coursework last year.

Choose two hypothesise that allow you to fully show off your statistical skills including some high level graphs like histograms and SRCC/standard deviation.

You don't need three, which a lot of people in my class did and it spread their efforts and made life difficult.

Then you just need to understand what you're doing, because if you don't then it's going to be hard to make graphs, analyse, etc.

At each stage, refer to the mark scheme and make sure you have addressed all the points.
What is SRCC?

Spearman's Rank? :s-smilie:
Reply 3
Original post by cannonball.
I got full marks on the stats coursework last year.

Choose two hypothesise that allow you to fully show off your statistical skills including some high level graphs like histograms and SRCC/standard deviation.

You don't need three, which a lot of people in my class did and it spread their efforts and made life difficult.

Then you just need to understand what you're doing, because if you don't then it's going to be hard to make graphs, analyse, etc.

At each stage, refer to the mark scheme and make sure you have addressed all the points.



What could a hypothesis be which's data can be done in 3 hours? About Film that is Please please please be online *fingers crossed(*
Original post by lizzie11
I need to write a plan for my coursework that I have already done, we are investigating how accurate people are at estimating the size/legnth of a line and angle! We asked year 7,9 and 11's in our school.

I'm aiming to get a high grade, I've dont SRCC, Standard Deviation, Scatter Diagrams and Box Plots so far.

I am so stuck on what to write, i need to include three hypothesis's, i've pretty much got three to write about, but i need help on how to structure it!

I need to know what to write as an intro and the rest of it, and how to structure it and how to end it!

Hope you can help haha!!


Hey I'm doing this coursework too, I've only just started it so I'm pretty short for time and I'm confused as to how to start...what sort of hypothesis did you use? I'm thinking of doing something with age and gender but need another interrelated hypothesis...any suggestions?
Reply 5
Original post by strawberries123
Hey I'm doing this coursework too, I've only just started it so I'm pretty short for time and I'm confused as to how to start...what sort of hypothesis did you use? I'm thinking of doing something with age and gender but need another interrelated hypothesis...any suggestions?


Hiya, I've finished this coursework and handed it in. I would do;

The year 11's will be more accurate than the year 7s. (could see this by the year 7's having a high standard deviation)

If a person overestimates on a line then they will overestimate on an angle (could show this by there being a positive correlation on the scatter graph with a srcc 0-1)

The year 7's estimations won't be normally distributed. (box plots, dot plots would show this)



In the plan start by explaining what you want to investigate and write all the questions you have down. Write in detail how you plan to collect your data and what sampling method you will use. Your plan needs to be detailed and explain exactly what your going to do and why you are doing this (what would it show?? why it is important to your investigation??)

Then do your graphs/calculaitons (i did about 5, spearmans rank, scatter diagram, standard deviation, dot plot and box plots) and write an interpretation after each and every one of them saying what they showed you and whether they prove or disprove your hypothesis.

For the evaluation explain what went well in your investigation, what you could have done better (taken a bigger sample as...) and for each hypothesis write whether they were proved or disproved in your investigation and say how you know this.

Its pretty simple and easy to understand when you know exactly what you need to do!! I would take a look at the mark scheme for every section to see what you need to do to get an A. The coursework is out of 40 marks and to get a C you need to get 20 marks, if you include standard deviation, box plots and spearmans rank then you are much more likely to get higher marks. Its important to use your own work and i would do whatever you think is necessary for your investigation and not just what i put.

Hope that helps you!! Message me if you need more help :smile:
Original post by lizzie11
Hiya, I've finished this coursework and handed it in. I would do;

The year 11's will be more accurate than the year 7s. (could see this by the year 7's having a high standard deviation)

If a person overestimates on a line then they will overestimate on an angle (could show this by there being a positive correlation on the scatter graph with a srcc 0-1)

The year 7's estimations won't be normally distributed. (box plots, dot plots would show this)



In the plan start by explaining what you want to investigate and write all the questions you have down. Write in detail how you plan to collect your data and what sampling method you will use. Your plan needs to be detailed and explain exactly what your going to do and why you are doing this (what would it show?? why it is important to your investigation??)

Then do your graphs/calculaitons (i did about 5, spearmans rank, scatter diagram, standard deviation, dot plot and box plots) and write an interpretation after each and every one of them saying what they showed you and whether they prove or disprove your hypothesis.

For the evaluation explain what went well in your investigation, what you could have done better (taken a bigger sample as...) and for each hypothesis write whether they were proved or disproved in your investigation and say how you know this.

Its pretty simple and easy to understand when you know exactly what you need to do!! I would take a look at the mark scheme for every section to see what you need to do to get an A. The coursework is out of 40 marks and to get a C you need to get 20 marks, if you include standard deviation, box plots and spearmans rank then you are much more likely to get higher marks. Its important to use your own work and i would do whatever you think is necessary for your investigation and not just what i put.

Hope that helps you!! Message me if you need more help :smile:


Will do, and thankyou!! :smile:
Reply 7
Hi I'm doing a similar stats coursework and I need to do pretty much the same thing as you guys and i'm really stuck!
I won't go into detail about it unless i Know someone can actually help but i'm really not sure about how to do a stratified sample for my data, if you could help please let me know i'd be really grateful!
(edited 10 years ago)
Reply 8
Original post by Lumos24520
Hi I'm doing a similar stats coursework and I need to do pretty much the same thing as you guys and i'm really stuck!
I won't go into detail about it unless i Know someone can actually help but i'm really not sure about how to do a stratified sample for my data, if you could help please let me know i'd be really grateful!


Hiya, well in your plan you should write about all the different sampling methods and how u carry them out and why you didn't use them (take to long, not suitable because you want to get an even mix of gender ect.) and then go on to stratified sampling, you should explain what stratified sampling is and then say why you used it for your investigation! Stratified sampling is good becauuse....

This is a good page on explaining stratified sampling:
http://www.bbc.co.uk/schools/gcsebitesize/maths/statistics/samplinghirev3.shtml

Read through the page and if you still don't understand how to carry out a stratified sample then just post on this post again :smile:
Reply 9
Original post by lizzie11
Hiya, well in your plan you should write about all the different sampling methods and how u carry them out and why you didn't use them (take to long, not suitable because you want to get an even mix of gender ect.) and then go on to stratified sampling, you should explain what stratified sampling is and then say why you used it for your investigation! Stratified sampling is good becauuse....

This is a good page on explaining stratified sampling:
http://www.bbc.co.uk/schools/gcsebitesize/maths/statistics/samplinghirev3.shtml

Read through the page and if you still don't understand how to carry out a stratified sample then just post on this post again :smile:


Hello again!! Thanks for commenting, i now fully understand what to do in my plan but i'm still a bit stuck on the stratified sampling so i'm hoping you can help me again!

My hypothesis is 'the further away a city is from London, the greater the Unemployment', i understand stratified sampling in general but not specifically for my hypothesis. My problem is that by doing the equation for stratified sampling it gives you the number of people you have to sample BUT on the website i got my data from only has the figures for unemployment in the cities in total, it doesn't have the individual people within those cities so i'm really stuck!! Have i got this wrong or can you not do a stratified sample from this data??

Again, thanks for you help! :smile:
(edited 10 years ago)
Original post by cannonball.
I got full marks on the stats coursework last year.

Choose two hypothesise that allow you to fully show off your statistical skills including some high level graphs like histograms and SRCC/standard deviation.

You don't need three, which a lot of people in my class did and it spread their efforts and made life difficult.

Then you just need to understand what you're doing, because if you don't then it's going to be hard to make graphs, analyse, etc.

At each stage, refer to the mark scheme and make sure you have addressed all the points.


How did you get full marks in this??? I'm finding it really difficult!! Any tips?? :smile:
Original post by lizzie11
Hiya, well in your plan you should write about all the different sampling methods and how u carry them out and why you didn't use them (take to long, not suitable because you want to get an even mix of gender ect.) and then go on to stratified sampling, you should explain what stratified sampling is and then say why you used it for your investigation! Stratified sampling is good becauuse....

This is a good page on explaining stratified sampling:
http://www.bbc.co.uk/schools/gcsebitesize/maths/statistics/samplinghirev3.shtml

Read through the page and if you still don't understand how to carry out a stratified sample then just post on this post again :smile:


OH MY GOD!!!! I understand how to do this now!!! Thanks so much!!
I am over the moon!!!!
:banana:
Reply 12
Original post by Lumos24520
OH MY GOD!!!! I understand how to do this now!!! Thanks so much!!
I am over the moon!!!!
:banana:


You're welcome!! I'm glad it helped you, I wish I had someone to tell me how to plan it when I was doing it, I was stuck for ages!!
Original post by lizzie11
You're welcome!! I'm glad it helped you, I wish I had someone to tell me how to plan it when I was doing it, I was stuck for ages!!

Yeah I guess but it's due in on Monday so I really need to get a move on!!
Original post by Lumos24520
Yeah I guess but it's due in on Monday so I really need to get a move on!!


By the way do you know whether you will lose marks if there's no correlation between your two variables?? Because that's happened to me and i'm really scared i will lose lots of marks for it :afraid::banghead:
Actually, does anyone know whether my coursework is wrong? There's no correlation between my two variable, i Can't change my hypothesis now it's WAY too late so what do i do?? Or is it okay as long as i identify it?? I'm so scared!! Someone help please!! :cry2:
Guys, i've got another question!! What is the best graph/diagram or calculation to show correlation between 2 sets of data? (except for scatter graphs and spearman's rank because i've done those already)
Original post by lizzie11
I need to write a plan for my coursework that I have already done


:rofl:
Original post by Lumos24520
Guys, i've got another question!! What is the best graph/diagram or calculation to show correlation between 2 sets of data? (except for scatter graphs and spearman's rank because i've done those already)


Do you think professional statisticians say "So I have produced strong evidence in support of my hypothesis now but I just need to find another two or three different representations of the same data as that will be extra convincing"?
Original post by Mr M
Do you think professional statisticians say "So I have produced strong evidence in support of my hypothesis now but I just need to find another two or three different representations of the same data as that will be extra convincing"?


I have this as well now :colonhash: coursework...

What else is there for bivariate data? Instead of spearmans rank and that... I will probably only use those but just to know.

Also, could I model data as a normal distribution?

Latest