How to Report Spearman Correlation: APA, Vancouver, and JAMA Formats

A correlation measures whether two things move together. Spearman's correlation, written with the Greek letter rho (ρ), is used when your data are not normally distributed or when they are ordered scores rather than true measurements. It is one of the most commonly reported statistics in clinical research — and one of the most commonly misformatted, which leads to avoidable corrections from journal editors. This guide gives you the exact format for the three style guides used in most medical journals.

To make it concrete, we will use a real example. In a diabetes screening study, researchers asked whether a patient's current fasting glucose level predicts their longer-term glucose control, measured as glycosylated haemoglobin (HbA1c).

What you need to report

A complete Spearman correlation report has five parts. The first is rho (ρ) itself, a number between -1 and +1. A value near +1 means the two things rise together; near -1 means one rises as the other falls; near 0 means almost no relationship. The second is the exact p-value — never simply "p < 0.05", always the real number. The third is n, the number of people the correlation was calculated from. The fourth is the direction, stated in words (positive or negative). The fifth is how strong the relationship is, described with a standard set of labels.

APA format

APA style writes the test as r_s(n) = .XX, p = .0XX. The small "s" marks it as Spearman rather than Pearson. APA removes the zero before the decimal point in both the correlation and the p-value. For example: "Stabilised blood glucose was positively correlated with glycosylated haemoglobin, r_s(390) = .53, p < .001." If the p-value is smaller than .001, write "p < .001", never "p = .000".

Vancouver format

Vancouver style — used by journals such as the Lancet, the NEJM, and the BMJ — keeps the leading zero and usually places the result in brackets within the sentence. For example: "A significant positive correlation was found between stabilised glucose and HbA1c (Spearman r = 0.53, p < 0.001, n = 390)."

JAMA format

JAMA style puts rho and the capital P in italics, removes the zero before the decimal in the p-value, but keeps it in the coefficient. For example: "Stabilised glucose and HbA1c were significantly correlated (Spearman ρ = 0.53, P < .001)."

How to describe the strength

Use a consistent set of labels. The widely used thresholds below are taken directly from Schober and colleagues, whose tutorial on correlation is the standard reference in clinical journals.¹

For example, the correlation between stabilised glucose and HbA1c was ρ = 0.53, placing it in the "strong" category.

Example data: Vanderbilt University Department of Biostatistics public teaching datasets (hbiostat.org/data). Figures computed with scipy from real data.

Common mistakes to avoid

Do not report r squared instead of r — that belongs to Pearson correlation, not Spearman. Do not write p = 0.000; write p < .001. Always say explicitly that the test is Spearman, not Pearson, so the reader knows which one you used. And always report n, because without it the reader cannot judge how reliable the correlation is.

What StatsPlease gives you

StatsPlease calculates Spearman's rho with the exact p-value, the sample size, a confidence interval, and a plain-language strength label. It produces the formatted sentence in both APA and Vancouver style, ready to paste into your results section.

References

1. Schober P, Boer C, Schwarte LA. Correlation Coefficients: Appropriate Use and Interpretation. Anesthesia & Analgesia. 2018;126(5):1763–1768. https://doi.org/10.1213/ANE.0000000000002864