When to use it
- •You have exactly two groups made up of different people.
- •Your outcome is continuous.
- •You want a simple mean comparison rather than a model with covariates.
Assumptions to check first
- •Independent observations — each participant appears in one group only.
- •Approximately normal outcome within each group (the test is fairly robust once groups are reasonably sized).
- •Homogeneity of variance — and if Levene's test flags a violation, report Welch's t instead rather than ignoring it.
What to report
- •Group means and standard deviations for both groups.
- •t with its degrees of freedom, and the exact p value.
- •An effect size — Cohen's d — with a confidence interval.
- •Whether you used Student's or Welch's version, if variances differed.
APA 7 example
Training participants scored higher (M = 78.4, SD = 9.1) than the control group (M = 71.2, SD = 10.3), t(118) = 4.05, p < .001, d = 0.74, 95% CI [0.37, 1.11], a medium-to-large effect.
Numbers are illustrative — the Advisor generates this sentence with your own results.
The mistake reviewers catch
Reporting p without an effect size says only that a difference exists, not whether it matters. With a large enough sample, a trivial difference will be "significant".
Run it on your own data
Paste or upload your dataset — nothing leaves your browser. The Analysis Advisor checks the assumptions above, runs t-test, and drafts the APA Methods and Results text with your numbers in it.
Open the Analysis Advisor →Related analyses
Frequently asked questions
When should I use t-test?
Do two separate groups differ on average? You have exactly two groups made up of different people.
What do I need to report for t-test in APA 7?
Group means and standard deviations for both groups. t with its degrees of freedom, and the exact p value. An effect size — Cohen's d — with a confidence interval. Whether you used Student's or Welch's version, if variances differed.
What is the most common mistake with t-test?
Reporting p without an effect size says only that a difference exists, not whether it matters. With a large enough sample, a trivial difference will be "significant".