Two variables that move together but not in a straight line, or an outcome that is ordinal rather than continuous. What ranking buys you, why you can compare it against Pearson without rerunning anything, and how to report it in APA 7.
Spearman's rho is Pearson's r computed on ranks rather than raw values. Each variable is ranked from lowest to highest, and the correlation is taken between the two sets of ranks.
That single change alters what is being tested. Pearson asks whether the relationship is a straight line. Spearman asks whether it is monotonic: whether one variable consistently rises as the other rises, regardless of whether it does so at a constant rate.
A relationship that curves upward steeply can therefore produce a modest Pearson r and a Spearman rho close to 1. The two disagreeing is informative rather than a problem.
In most software, running Spearman means selecting it before you see anything, which forces the decision at exactly the moment you know least about the relationship.
Quanta reports Pearson, Spearman, and Kendall tau-b in one table, each with its own p value and sample size, alongside a scatter with a least-squares fit. The comparison is the point.
Read the gap. Rho close to r means the relationship is essentially linear and the conventional Pearson report is defensible. Rho materially higher than r means the association is real but curved, and that is a substantive finding about your variables rather than a technical choice about coefficients.
This also removes a quiet source of bias. When choosing a coefficient requires a rerun, there is a pull toward keeping whichever came out larger. Seeing all three at once makes that visible to you and reportable to a reader.
Ordinal data. A single Likert item, an education level, a pain rating, a ranked preference. Ranking is what the data already is, so nothing is discarded.
Monotonic but curved relationships, where Pearson would understate a real association.
Outlier-prone variables. A rank-based coefficient limits how far one extreme value can distort the result, since the largest value is only ever one rank above the second largest.
Where there are many tied values, particularly on a short scale, Kendall's tau-b handles ties more gracefully.
Bring in the data. Import from CSV, Excel, SPSS .sav, Stata .dta, or a ReliCheck survey package.
Configure the two fields on the right and all three coefficients appear together in the center panel with the scatter beneath them.
Upload ordinal categories as text, not as numeric codes. If an education level arrives as 1, 2, 3, Quanta reads it as a number. Bringing categories in as labels keeps the variable typed correctly.
Use the action bar to add the result to a report, copy the table, or copy the APA text.
Report rho with its sample size and p, and name the coefficient explicitly.
Reported workload was positively associated with burnout, rs(248) = .38, p < .001.
State why you chose Spearman. One clause is enough: because the outcome was ordinal, or because the relationship was monotonic but not linear.
Where Pearson and Spearman differ meaningfully, report both and say what the difference indicates. It is more informative than either number alone.
Do not interpret rho as a proportion of shared variance. Squaring a rank correlation does not carry the meaning that squaring Pearson's r does.
Every result carries a ReliCheck Intelligence card with one action: explain this result in plain language. It runs entirely on your Mac with Apple Intelligence. No upload, no API key, nothing to configure. The model only explains the result Quanta has already computed and validated. It never computes a statistic, and it never applies a decision on your behalf.
From the action bar you can add the result to a report, copy the table, or copy the APA text. Exported reports carry the APA table, its note, and a caution specific to that analysis beneath it.
Quanta's correlation and nonparametric procedures are validated against R's cor and against hand-computed closed-form values, matched exactly. The engine is pure Swift with no R or Python at runtime. Full record on the validation page.
Correlation is part of every Quanta subscription, $9.99 a month or $79.99 a year after the 30-day trial. One price for everyone. See pricing.
The analysis, the report builder, and the on-device explanations all work with the network off. Your dataset is never uploaded.
Pearson for linear relationships between continuous variables · Kendall's tau-b for small samples with many ties · Partial correlation holding a third variable constant · the complete list on the analyses page.