You are Here: Home >< Maths

# Spearmans rank correlation watch

1. Hey im currently doing a statistics project on supermarkets and i want to compare how cheap some items are with how many people shop at the supermarkets (to see if cheapness is the reason they are shopping there.) When it comes to ranking them do i still do 1 or 5 as the lowest FOR BOTH because surely if i rank the cheapest set of items at a supermarket as 1 and rank least amount of people who shop at one supermarket as 1 surely this doesnt work?

Hope you understand what im trying to say
Anyway if I do both as 1 it comes out around -0.5 but if i do one as 1 and one as 5 it comes out as +0.5(which makes sense as more people are likely to go to a supermarket that's cheaper)

thanks
2. (Original post by Robpattinsonxxx)
Hey im currently doing a statistics project on supermarkets and i want to compare how cheap some items are with how many people shop at the supermarkets (to see if cheapness is the reason they are shopping there.) When it comes to ranking them do i still do 1 or 5 as the lowest FOR BOTH because surely if i rank the cheapest set of items at a supermarket as 1 and rank least amount of people who shop at one supermarket as 1 surely this doesnt work?

Hope you understand what im trying to say
Anyway if I do both as 1 it comes out around -0.5 but if i do one as 1 and one as 5 it comes out as +0.5(which makes sense as more people are likely to go to a supermarket that's cheaper)

thanks
Both ways are showing the same result, so it doesn't really matter which way around you do it. The important thing is the interpretation of your results. To put it crudely, you're either showing a positive correlation between cheapness and visits, or a negative correlation between expensiveness and visits. Can you see why these are the same?
3. Agree ^ As long as you are clear in your mind as to what you result shows then its fine and it does work. The -0.5 shows medium negative correlation between the high cost of products and visits and the +0.5 shows medium positive correltation between the low cost of products and number of visits. As mentioned already, this is the same thing, just looked at in two ways.
4. Thanks guys
5. ASDA TESCO Sainsburys Morrisons Somerfield
Price of 10 branded food and 10 non branded food items £23.78 £24.77 £25.58
£24.65 £27.31

Number of people who go there for their weekly shop 16 6 3 2 2
Rank 1 1 3 4 2 5
Rank 2 5 4 3 1.5 1.5
d 4 1 1 0.5 3.5
D(squared) 16 1 1 0.25 12.25 30.5

N=5
D(squared)=30.5

Sorry hope you don't mind. The answer is -0.525. I'm not looking for the answer particularly but what would you say with this result? Just the obvious? That more people go to the supermarkets because there cheaper? Because I don't know if it follows the actual data I have. Do you understand what I mean?
6. By ranking the highest number in each criteria as 1, i get Rs as -0.225, here's how:

price rank ¦ 5 ¦3 ¦ 2 ¦ 4 ¦ 1 ¦
Number rank ¦1 ¦ 2 ¦ 3 ¦ 1.5 ¦ 1.5 ¦
d ¦ 4 ¦ 1 ¦ 1 ¦ 2.5 * ¦ 0.5 ¦ * think this is your mistake, looks like you have 3.5
d^2 ¦ 16 ¦ 1 ¦ 1 ¦ 6.25 ¦ 0.25 ¦
sum of d^2 = 24.5

Rs = 1 - [ (6* sum of d^2)/((n^3)-n)] = 1 - [ (6*24.5)/(125-5)] = 1 - 1.225 = -0.225

I'm not meaning this to be patronising, so sorry if it looks that way. Anyway, this result shows a fairly weak negative correlation between the number of visitors and 'expensiveness'. Hope this helps.
7. Thanks so would the conclusion be that there is a correlation between them or not because that's what I'm confused about seeing it is negative.
8. Just becuase it's negative doesn't mean there can be no correlation although in this case it is fairly weak. +1 is strong positive correlation, -1 is strong negative correlation and the closer to 0, the correlation gets weaker if this makes sense.

### Related university courses

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: February 5, 2010
Today on TSR

### Top unis in Clearing

Tons of places at all these high-ranking unis

Poll
Useful resources

### Maths Forum posting guidelines

Not sure where to post? Read the updated guidelines here

### How to use LaTex

Writing equations the easy way

### Study habits of A* students

Top tips from students who have already aced their exams