Short definition
Spearman rank correlation — a statistical measure of how closely two rankings agree, from −1 (opposite order) through 0 (no monotonic association) to +1 (the same order).
Also known as: Spearman's rho, rank correlation, Spearman coefficient.
Spearman rank correlation is a statistical measure of how strongly the orderings of two ranked lists correspond, expressed as a single number between −1 and +1. We use it to check whether things like speed scores line up with search positions.
What Spearman rank correlation is
Spearman rank correlation, usually written as ρ (rho) or rs, measures the association between two sets of ranks. For example, you rank nineteen web development agencies by their PageSpeed scores and rank the same nineteen by their Google position numbers, assigning average ranks to tied values. You then calculate the correlation between those ranks. Matching orders produce a value near +1; opposite orders produce a value near −1. A value near 0 indicates little monotonic association: no consistent tendency for one value to rise or fall as the other rises. Other kinds of relationships may still exist.
The difference from Pearson correlation is that Spearman uses the order, not the size of the gaps between the underlying values. Google positions 1 and 2 may attract very different numbers of clicks, but Spearman treats them as adjacent ranks. Using ranks makes it less sensitive to extreme values, though it does not remove sampling bias or other limitations. Wikipedia explains the calculation.
How to read the number
As a rough guide, values around ±0.8 or stronger indicate a strong association; values with a magnitude between 0.4 and 0.7 are often described as moderate; values closer to zero suggest a weak association or none. These labels are conventions, not significance tests. With a small sample of around twenty pages, an apparently strong result can still arise by chance. Always report the sample size and specify what you compared.
More importantly, correlation does not establish cause. A third factor could affect both variables: a strong brand might be associated with both high rankings and a polished website. Treating that association as proof that one causes the other goes beyond what the number supports.
Why we use it
In our lab and field data research, we asked whether higher PageSpeed scores were associated with better rankings for a competitive query. Spearman rank correlation lets us compare those orders without assuming a linear relationship between the scores and position numbers.
The September 2026 result was +0.32 between lab score and position number across 19 agencies: higher scores tended to go with slightly worse rankings.
The Chrome UX Report (CrUX) provided loading-performance data from an eligible subset of Chrome users. CrUX LCP (p75) had a correlation of −0.07 across the seven sites with current field observations. That is close to zero, not proof that speed has no effect. The separate History API dataset is not part of that seven-site calculation. Page Authority had a stronger association with position in this study, but none of these correlations establishes what caused the rankings.