Ranks and Rank Sums in Nonparametric Statistics constitute a foundational methodological framework within inferential statistics that utilizes ordinal data transformation rather than raw metric values for analysis. The core principle involves assigning integer positions to observations based on their magnitude relative to one another, thereby reducing the influence of outliers and non-normal distributions while preserving the underlying rank-order information. This domain-specific mechanism serves as a robust alternative to parametric tests by relying exclusively on distribution-free properties, ensuring validity under conditions where assumptions regarding population normality or interval scaling are not met.
Prerequisite chain context: requires Ordinal Variable Classification in Statistics.