Turn on thread page Beta
 You are Here: Home

# spearman's ramk watch

1. hi,
i am currently doing my gcse coursework in year10 for Geography. I have almost completed it, however, i have to use spearman's rank to compare the ethnic minority of Wellinborough (my home town). I have recieved the information but i do not know how to use spearman's rank.
hope to have a reply soon.
Ian.
2. ...

what are you asking for? help to use the spearman's rank?

well first of all, you need some data to compare yours to. then im not sure what to do; havent done it in soo long.
3. (Original post by ianbradshaw2002)
hi,
i am currently doing my gcse coursework in year10 for Geography. I have almost completed it, however, i have to use spearman's rank to compare the ethnic minority of Wellinborough (my home town). I have recieved the information but i do not know how to use spearman's rank.
hope to have a reply soon.
Ian.
I don't quite see how you could your SRCC here. What is the exact question and I will try to help (I have GCSE Maths, Statistics, A Level Maths, A Level Further Maths (all at the top grade), and I am doing maths as part of my Cambridge course).
4. (Original post by kikzen)
...

what are you asking for? help to use the spearman's rank?

well first of all, you need some data to compare yours to. then im not sure what to do; havent done it in soo long.
OK. Here is a generic example. Suppose you want to calculate SRCC for the following two sets of data (I will use the example of say, age and IQ).

Age..........................20. .... 25.......30.......35.......40... ....45
Ave IQ.....................108.....1 10.....114.....118.....120...... 118
Age Rank...................1........ .2.........3........4.........5. ........6
IQ Rank.....................1...... ...2.........3........4.5....... 6.......4.5
difference in rank (d)...0.........0.........0..... ...-0.5......-1......1.5
d^2..........................0.. .......0.........0........0.25.. ...1.......2.25
Sum (d^2) = 3.5

What I did above was to write out the data as a table and assign ranks to the age and the IQ, from lowest to highest. Where to raw data values are the same, you need to take the average of the next two ranks. In the case above, the IQ ranks were 4.5 as the average of the 4th and 5th ranks were taken.

Next you need to find the difference in ranks which is calculated by subtracting one row from the other. Then this difference is squared. Finally the sum of the squares of differences is calculated.

Spearman's Rank Correlation Coefficient (r) is calculated by using the following formula:

r = 1 - 6.Sum(d^2)
..........n((n^2)-1)

Where sum(d^2) is the value calculated above, and n is the number of pairs of data, in this case 6.

So the value of r for my imaginary data is 1 - 21
................................ .............................6*3 5

= 1 - 0.1
=0.9

This value is close to 1 so it corresponds to a strong positive correlation. This means that if the data was plotted on a scatter graph, the line of best fit would have a positive gradient (positive correlation), and the data points would generally be very close to it (strong correlation).

I hope this helps. I apologise for the dots in the table, they are for spacing purposes.
5. (Original post by ianbradshaw2002)
hi,
i am currently doing my gcse coursework in year10 for Geography. I have almost completed it, however, i have to use spearman's rank to compare the ethnic minority of Wellinborough (my home town). I have recieved the information but i do not know how to use spearman's rank.
hope to have a reply soon.
Ian.
See if your two sets of data are normally distributed first. If they are, it's unecessary to use Spearman's Rank, you can use Pearson's, which is given by:

r=

(1/n)*Sum(xi-xbar)(yi-ybar)
________________________________ ___________________

sqrt((1/n)Sum(xi^2)-xbar^2) * sqrt((1/n)Sum(yi^2)-ybar^2)
6. (Original post by Alec)
See if your two sets of data are normally distributed first. If they are, it's unecessary to use Spearman's Rank, you can use Pearson's, which is given by:

r=

(1/n)*Sum(xi-xbar)(yi-ybar)
________________________________ ___________________

sqrt((1/n)Sum(xi^2)-xbar^2) * sqrt((1/n)Sum(yi^2)-ybar^2)
Remember that this is for a GCSE coursework, testing that the data sets fit a N-dis is a little too high a level I think (AS Further).
7. (Original post by chrisbphd)
I don't quite see how you could your SRCC here. What is the exact question and I will try to help (I have GCSE Maths, Statistics, A Level Maths, A Level Further Maths (all at the top grade), and I am doing maths as part of my Cambridge course).
I have changed my idea and i am finding out 'if there is a relationship between the percentage of non-white populations and the percentage of rented in house' usinhg spearmans rank.
It is possible because my brother did the same thing and he progressed to get an A*. and seeing that this is the only coursework i do in Geography, i am putting all my effort to get it right. I have asked my brother to show me, however, he can not explain it.
Now can you help? All i know is that i have to show my results in a scattergraph to see if there is a correlation...
(i am going to get my information from the neighbourhood statistics website. 2001 census data..)
hope for a reply soon.
Ian.
thanks for your response earlier.
8. Sorry to hijack your thread slightly, but...

If your data is normally distributed you use Pearson's product moment correlation coefficient, if not then you use Spearman's... but why is that?

Why does normality imply Pearson's should be used, and abnoramlity imply Spearman's should be used?

Thanks.
9. i had the same notion but later found that pearsons product moment coefficiency can only be used with interval or ratio data where spearmans rank deals with ordinal data.

Turn on thread page Beta
TSR Support Team

We have a brilliant team of more than 60 Support Team members looking after discussions on The Student Room, helping to make it a fun, safe and useful place to hang out.

This forum is supported by:
Updated: December 10, 2003
Today on TSR

### He lied about his age

Thought he was 19... really he's 14

Poll
Useful resources

## Articles:

Debate and current affairs forum guidelines