Reporting an independent samples t-test (APA 7)

A complete report states the test, both group means and standard deviations, the t statistic with its degrees of freedom, the exact p-value, the 95% confidence interval of the mean difference and an effect size.

Template

An independent samples t-test compared [outcome] between [group A] (M = x.xx, SD = x.xx) and [group B] (M = x.xx, SD = x.xx). The difference was [direction], t(df) = x.xx, p = .xxx, 95% CI [x.xx, x.xx], d = x.xx.

Rules that are often missed

  • Report p to three decimals without a leading zero (p = .025); below .001 write p < .001.
  • Use Welch's t-test by default; its degrees of freedom are usually not whole numbers (e.g. t(17.78)).
  • Give the confidence interval of the difference, not of each mean.
  • Italicise statistical symbols (t, p, M, SD, d) in your manuscript.

When not to use it

If the outcome is clearly non-normal in small groups, use the Mann–Whitney U test and report medians and the rank-biserial correlation instead.

Frequently asked questions

Why Welch instead of Student's t-test?

Welch's test does not assume equal variances and loses little power when they are equal, so it is the safer default.

What is a small, medium or large Cohen's d?

Conventionally 0.2, 0.5 and 0.8, but judge it against effects that matter in your field.

Try it with your data See a sample dashboard

Related