How to Report Correlation Results in APA 7 Format (Templates and a Worked Example)

To report correlation results in APA 7 format, state the direction and strength of the relationship in plain words, then give the italicized coefficient with degrees of freedom, the exact p value, and a 95% confidence interval. For example: r(48) = .38, p = .007, 95% CI [.11, .60]. Most students get the number right and the formatting wrong. Common errors include leaving out degrees of freedom, adding leading zeros, and forgetting italics, and these are exactly what supervisors and reviewers flag.

Key Takeaways

  • An APA 7 correlation follows the template r(df) = .XX, p = .XXX, where degrees of freedom equal the sample size minus 2 (N − 2).
  • Correlation coefficients and p values never take a leading zero in APA 7 because neither can exceed 1 (write .38, not 0.38).
  • Statistical symbols such as r, rs, and p are italicized, while numbers and brackets are not.
  • APA 7 recommends reporting a 95% confidence interval for the correlation coefficient alongside the p value.
  • Report exact p values to two or three decimals, except when p is below .001, which is written as p < .001.

What does a correlation tell you?

A correlation measures how closely two variables move together. If students who sleep more tend to score higher on exams, the correlation is positive. If more screen time goes with lower grades, it is negative. The coefficient runs from −1 (perfect negative relationship) through 0 (no relationship) to +1 (perfect positive relationship).

Two tests cover most thesis work. Pearson's r suits two continuous, roughly normally distributed variables with a linear relationship. Spearman's rho (rs) suits ranked data, ordinal scales, skewed distributions, or relationships that are monotonic but not linear. If you are unsure which applies to your data, see our guide on Pearson vs Spearman correlation.

What is the APA 7 format for a Pearson correlation?

The APA 7 template for a Pearson correlation is:

r(df) = .XX, p = .XXX, 95% CI [LL, UL]

Each element has a specific rule:

  • r is italicized.
  • df goes in parentheses directly after r, with no space. For a correlation, df = N − 2.
  • The coefficient has two decimal places and no leading zero. Negative values use a minus sign: r(48) = −.27.
  • p is italicized, has no leading zero, and is reported exactly (p = .007, not p < .05). Values below .001 are written p < .001.
  • The 95% CI goes in square brackets with the lower and upper limits separated by a comma.

How do you report a Spearman correlation in APA 7?

A Spearman correlation follows the same structure, with rs in place of r:

rs(48) = .35, p = .013

Some journals use the Greek letter ρ instead. Either is acceptable, but use one symbol consistently throughout your manuscript.

Pearson vs Spearman: key reporting differences

Feature Pearson's r Spearman's rs
Data type Continuous, interval/ratio Ordinal, ranked, or skewed continuous
Relationship assessed Linear Monotonic
Symbol in APA 7 r rs (or ρ)
Degrees of freedom N − 2 N − 2
Example write-up r(48) = .38, p = .007 rs(48) = .35, p = .013
Assumption to mention Normality, linearity None beyond monotonicity

Worked example: reporting a correlation step by step

Suppose you measured average nightly sleep and final exam scores for 50 undergraduates and ran a Pearson correlation. Your software returns r = .38, p = .0065, N = 50.

Step 1: Calculate degrees of freedom. df = 50 − 2 = 48.

Step 2: Format the coefficient. .38 has two decimals and no leading zero.

Step 3: Format the p value. .0065 rounds to .007, which is below .05, so the result is statistically significant.

Step 4: Add the confidence interval. Using Fisher's z transformation, the 95% CI is [.11, .60]. Because the interval excludes zero, it supports the significant result. The interval is also wide, which shows the estimate is imprecise with 50 participants.

Step 5: Interpret the strength. Cohen's (1988) conventional benchmarks treat |r| = .10 as small, .30 as medium, and .50 as large. A value of .38 is a moderate correlation.

Step 6: Write the sentence.

A Pearson correlation showed a significant, moderate positive relationship between average nightly sleep and final exam scores, r(48) = .38, p = .007, 95% CI [.11, .60]. Students who slept more tended to score higher.

Notice what the sentence does not say: that sleep causes higher scores. A correlation describes an association, and your write-up should never claim causation.

How do you report a non-significant correlation?

A non-significant correlation is reported in exactly the same format, and the wording should not overstate the result. For example: "There was no significant correlation between caffeine intake and exam scores, r(48) = .12, p = .406, 95% CI [−.16, .38]." Do not write that the variables are "unrelated." A non-significant result means the data could not rule out a zero correlation, and a CI that spans from negative to positive values shows that uncertainty honestly.

How do you report multiple correlations?

When you have more than three or four correlations, APA 7 recommends a correlation matrix table rather than a long paragraph. Put the variables in the rows and columns, fill only the lower triangle, and mark significance with asterisks defined in a table note (e.g., *p < .05, **p < .01). Include means and standard deviations in the first columns, then summarize the key relationships in the text.

Can software write the APA correlation report for you?

Formatting by hand is where most errors creep in. The usual mistakes are typing 0.38, forgetting df, or rounding p inconsistently. Traditional packages such as SPSS, Stata, and R give you the raw numbers but not a formatted sentence. SPSS also costs roughly $100 per month on an individual subscription. JASP and jamovi are free and produce cleaner tables, but you still have to write the prose yourself.

StatRyx is an AI-powered statistical analysis tool that replaces manual SPSS workflows with automated, APA 7-formatted reporting. You upload your dataset and choose two variables. StatRyx checks normality to recommend Pearson or Spearman, then returns a ready-to-paste sentence with df, exact p, and the 95% CI already formatted. If you are still deciding which test fits your design, StatRyx also explains why it chose that test, so you can defend the choice in your methods section.

Stop calculating this by hand — run it free in StatRyx → Try StatRyx

Frequently Asked Questions

Do you put a zero before the decimal in a correlation in APA 7?

No. APA 7 drops the leading zero for any statistic that cannot exceed 1, which includes correlation coefficients and p values. Write r = .38 and p = .007, not 0.38 and 0.007.

How do you calculate degrees of freedom for a correlation?

Degrees of freedom for a Pearson or Spearman correlation equal the sample size minus 2. With 50 participants, df = 48, reported as r(48). If cases have missing data, use the number of complete pairs actually analysed.

What counts as a strong correlation?

Cohen's widely used conventions treat |r| around .10 as small, .30 as medium, and .50 or above as large. These are rough guides, not rules. A correlation of .20 can matter in medical research, while .60 may be expected between two versions of the same questionnaire.

Should I report r squared for a correlation in APA 7?

It is optional but often helpful. Squaring r gives the proportion of shared variance; for example, r = .38 means about 14% shared variance. APA 7 does not require it, but including it can make the practical size of the relationship clearer to readers.

Is it okay to write p < .05 instead of the exact value?

APA 7 asks for exact p values to two or three decimals, such as p = .007. The only exception is very small values, which are written p < .001. Reporting only p < .05 hides information and is likely to be corrected by reviewers.

Stop calculating this by hand. Upload your dataset and StatRyx's AI runs the correct test and returns copy-paste-ready APA 7 output in seconds — no SPSS license, no syntax.

Run your data through StatRyx free →
← All posts