Spearman's Rho and Non-Monotone Data

Question

I have used SPSS to calculate Spearman's Rho for ordinal and scale data. No significant correlations were found. However, I am not sure that my data is monotonic and I do not understand how to interpret the scatter plots to determine if the data is monotonic. Can I still correctly assert that no significant correlations exist in the data?

Thank you.

Answer 1

Spearman’s Rho is a test of monotonicity.

So, if the p-value associated with the Spearman’s Rho you calculated is < .05, then you have a significant monotonic relationship. In your case, since Spearman’s Rho is not significant, you cannot claim a monotonic relationship.

The scatter plot is interpreted similarly to when you have a Pearson correlation, EXCEPT the Spearman Rho does not have to approximate a line (it could be curvilinear for example). Either both variables increase together (positive relationship) or one variable increases while the other decreases (negative relationship). The difference is that the data does not have to approximate the line for a monotonic relationship (a line ear relationship is not required).

Hope that helps…

Joseph