# How to count Spearman Rank order correlation

1. May 8, 2013

### Drudge

1. The problem statement, all variables and given/known data
calculate the rank order correlation between the following data:

6, 5, 4, 2, 3, 3, 8, 3, 7, 6, 7, 5, 5, 4, 2, 7, 6, 2, 4, 6

4, 3, 6, 7, 6, 7, 1, 9, 1, 2, 3, 4, 5, 5, 7, 1, 2, 9, 5, 4

2. Relevant equations

Following the output from http://www.vassarstats.net/corr_rank.html, given the ranks, then from each individual rank subtract the equivalent opposing / matching rank. Lastly, raise every subtraction to the power of two and sum up results:

Ʃ( #x$_{i}$ - #y$_{i}$ )$^{2}$

3. The attempt at a solution

I have used the aforementioned online site to help my calculations, and have used the ranks given there, from the appropriate variables, to perform the previously described actions.

The answer is ≈ -0.93. This is confirmed by vassarstat.net. However, vassarstat does not give any explanation how to perform the rest of the problem.

I can only get ≈-0.90. After Ʃ( #x$_{i}$ - #y$_{i}$ )$^{2}$, I get 2529. Then I use the spearman rank order correlation equation 1 - $\frac{6 * 2529}{20 * (20^{2} - 1)}$ ≈ -0.90

2. May 9, 2013

### Drudge

Anyone? Exams are just behind the corner and I can not figure this one out. Hate it if they would ask to use this in the test. It could be so simple if i could just understand