Spearman's rho - square it?

[jmw]

When using Pearson's correlation one can take the coefficient (R) and square it (R2) to get the coefficient of determination, which is interpretated as the fraction of variance shared between the two variables.

My question is: is there any meaning or value in doing the corresponding thing with the Spearman rank correlation coefficient (Rs) when using Spearman rank correlation?

Can Rs2 be interpretated in any meaningful way?

[omega]

Spearman rank correlation coefficient is non-parametric. It looks at ranks, not the data directly. So I don't think Rs^2 has any meaning. It contains rank squared, and cannot be interpreted.