APA REPORTING GUIDE
How to report a t-test in APA style
An APA t-test report states which t-test you ran, the means and standard deviations, t with its degrees of freedom, the exact p value, and an effect size such as Cohen's d.Explore research statistics & data analysis →FORMULA OR CORE IDEA
t(df) = t value, p = .XXX, d = X.XXWORKED EXAMPLE
An independent-samples t-test showed that students who used practice quizzes (M = 82.3, SD = 7.4) scored significantly higher than those who did not (M = 76.1, SD = 8.9), t(58) = 2.94, p = .005, d = 0.76.QUICK REFERENCE
APA t-test templates
Replace the numbers with your own results.| Test | Template | df rule |
|---|---|---|
| Independent-samples t-test | t(58) = 2.94, p = .005, d = 0.76 | n₁ + n₂ − 2 |
| Welch's t-test | t(41.7) = 2.35, p = .024, d = 0.62 | Welch-Satterthwaite (decimal) |
| Paired-samples t-test | t(29) = 4.12, p < .001, d = 0.75 | n − 1 (pairs) |
| One-sample t-test | t(24) = −2.18, p = .039, d = 0.44 | n − 1 |
| Mann-Whitney U (non-parametric) | U = 312, z = −2.41, p = .016, r = .31 | Report medians |
| Wilcoxon signed-rank (non-parametric) | z = −3.02, p = .003, r = .55 | Report medians |
STEP BY STEP
Say which t-test you used
Independent-samples (two separate groups), paired-samples (same people twice), or one-sample (one group against a fixed value). For unequal variances, say Welch's t-test.
Report the means and standard deviations
Give M and SD for each group or time point so the direction of the difference is clear.
Give t with its degrees of freedom
Independent t-test: df = n₁ + n₂ − 2. Paired or one-sample: df = n − 1. Welch's t-test has a decimal df, such as t(41.7) = 2.35.
Add the exact p value
Write p = .005, not p < .05. Use p < .001 only for very small values, and drop the leading zero.
Report an effect size
Cohen's d: about 0.2 small, 0.5 medium, 0.8 large. A 95% confidence interval for the mean difference is also recommended by APA.
Interpret in plain words
Finish with a sentence that states what the result means for your research question, without overstating causation.
COMMON QUESTIONS
Frequently asked questions
Do I report a negative t value?
Yes, report the sign your software gives, but make the direction clear in words. Many authors order the groups so the larger mean comes first and t is positive.
Should I report a one-tailed or two-tailed p value?
Report two-tailed p values unless you pre-registered a directional hypothesis. If you use a one-tailed test, say so explicitly.
How many decimal places should I use?
APA recommends two decimal places for t, d, M, and SD, and two or three for p values (three is common, for example p = .005).
What if my data are not normal?
Check with a Shapiro-Wilk test. If normality is clearly violated in small samples, use the Mann-Whitney U test (independent) or the Wilcoxon signed-rank test (paired) and report medians.