APA 7 How to Report Statistics Results: A Practical Guide for Researchers

Your supervisor sent the draft back with "fix APA formatting throughout" scrawled in the margin — and you have no idea whether p should be italicized, how many decimals to use, or where to put that confidence interval. If you're searching "APA 7 how to report statistics results," you're probably staring at a results section that looks technically correct but feels wrong.

This guide gives you the actual rules from the APA 7th edition Publication Manual, with worked examples you can copy and adapt. No filler, no vague advice — just the patterns reviewers expect to see.

The Core Rules of APA 7 Statistical Reporting

Before getting into specific tests, three formatting rules apply to nearly everything you report:

1. Italicize statistical symbols that use Latin letters: t, F, p, r, M, SD, N, df. Greek letters (η², χ²) are not italicized.
2. Use two decimal places for most statistics. Use three for p values (e.g., p = .003). For p < .001, write exactly that.
3. Drop leading zeros for numbers that cannot exceed 1: p, r, β, η². Keep leading zeros for everything else (M = 0.45 is fine; p = 0.03 is not — it should be p = .03).

A quick note on p values: APA 7 prefers exact p values (e.g., p = .024) rather than the older p < .05 convention, unless the value is below .001.

How to Report a t-Test in APA 7

The format is: t(df) = value, p = value, d = value, 95% CI [lower, upper].

Worked example: You compare exam scores between 30 students who used flashcards (M = 82.3, SD = 6.1) and 30 who used rereading (M = 76.8, SD = 7.4). Here's the proper write-up:

Students using flashcards scored significantly higher (M = 82.30, SD = 6.10) than students using rereading (M = 76.80, SD = 7.40), t(58) = 3.14, p = .003, d = 0.81, 95% CI [2.00, 9.00].

Notice what's included: the degrees of freedom in parentheses, the exact p value, Cohen's d as the effect size, and a confidence interval for the mean difference. APA 7 increasingly expects effect sizes and CIs — running this in StatRyx automatically generates all four, so you don't have to compute d by hand.

How to Report ANOVA Results

The format is: F(df between, df within) = value, p = value, η² = value (or η²p for partial eta-squared).

Worked example: In a study of 45 psychology students randomized to three study conditions, you find a significant effect of condition on memory recall:

A one-way ANOVA revealed a significant effect of study condition on recall accuracy, F(2, 42) = 4.31, p = .019, η² = .17.

Let's break that down: F(2, 42) means 2 numerator degrees of freedom (3 groups − 1) and 42 denominator degrees of freedom (45 − 3). The η² of .17 indicates 17% of variance in recall is explained by study condition — a large effect by Cohen's conventions. Always follow a significant omnibus F with post-hoc tests (Tukey HSD, Bonferroni) and report those pairwise comparisons too.

How to Report Regression and Correlation

For correlations: r(df) = value, p = value.

Sleep hours and exam performance were positively correlated, r(48) = .42, p = .002.

For regression, report the overall model and each predictor:

The model significantly predicted job satisfaction, F(3, 96) = 12.45, p < .001, R² = .28. Autonomy was the strongest predictor (β = .41, p < .001), followed by salary (β = .22, p = .021).

How to Report Chi-Square

The format is: χ²(df, N = sample size) = value, p = value.

The association between gender and program choice was significant, χ²(2, N = 120) = 8.94, p = .011, Cramér's V = .27.

The sample size goes inside the parentheses with df — this is a quirk specific to chi-square that catches a lot of students out.

Quick Reference Table: APA 7 Reporting by Test

Run this analysis free in StatRyx → Try StatRyx