What information is lost when interval data are changed into ranks for Spearman correlation?

When interval data are transformed into ranks for use in a Spearman rank-order correlation coefficient analysis, the primary information lost is the precise magnitude of the differences between the data points.

While interval scales provide information about the specific amount of difference between values, ordinal or rank scales only indicate the relative order of quality or quantity. By converting interval data to ranks, you discard the quantitative distance between observations, retaining only their relative position in the sequence. Consequently, the analysis can no longer account for the exact degree of difference between values, as the scale no longer reflects the interval nature of the original data.12