Which correlation coefficient is based on ranked values rather than raw data?

Spearman's rho is a statistical measure that assesses how well the relationship between two variables can be described using a monotonic function. It is based on ranked values, which means that it evaluates the degree of association between two datasets by first converting their raw scores into ranks. This approach is particularly useful when the data does not meet the assumptions of normality required for Pearson's correlation coefficient, making Spearman's rho more robust to outliers and nonlinear relationships.

In contrast, Pearson's correlation coefficient relies on raw data and measures linear correlation, while autocorrelation examines the correlation of a signal with a delayed copy of itself, and the coefficient of determination (R-squared) quantifies the proportion of variance in the dependent variable that can be predicted from the independent variable(s). These other measures do not utilize ranked values, which is what makes Spearman's rho distinct in its methodology.