Why Reporting Your t-Test Correctly Matters
You ran the analysis, SPSS gave you the numbers, and your result is significant — congratulations! But before you celebrate, one more job remains: writing those results up in APA style. This is the step where many students lose marks, not because their analysis was wrong, but because their reporting was sloppy.
Supervisors, journal reviewers, and examiners all read the results section with one question in mind: did this researcher report enough detail for me to trust — and repeat — what they did? A cleanly reported t-test answers that question in a single sentence. In this guide, you will learn the exact APA 7th edition formatting rules, follow a five-step writing process, see a fully worked example for each of the three t-test types, copy ready-made templates, and avoid the six mistakes students make most often.
The APA Formula for a t-Test: What Goes In
Every APA-style t-test report contains the same building blocks. Think of it as a sentence with a fixed skeleton:
t(df) = t-value, p = p-value
Here is what each piece means and how to format it:
| Component | What to write | Formatting rule |
|---|---|---|
| The statistic | Lowercase italic t | Italicize the letter only — never the numbers or parentheses |
| Degrees of freedom | In parentheses right after t, e.g., t(36) | No spaces inside the parentheses |
| The t value | Two decimal places, e.g., 2.56 | Keep the leading zero (values can exceed 1) |
| The p value | Exact to 2–3 decimals, e.g., p = .015 | No leading zero (p cannot exceed 1); write p < .001 for very small values, never p = .000 |
| Group descriptives | M = 78.50, SD = 8.20 for each group | Italicize M and SD; keep leading zeros |
| Effect size | Cohen’s d = 0.83 | Recommended by APA 7th edition alongside every test |
| Confidence interval | 95% CI [1.51, 12.90] | Lower bound first, in square brackets |
Notice the pattern: italicize statistical symbols (t, p, M, SD, d) but not the numbers, parentheses, or words like “CI”. This one habit instantly makes your reporting look professional.
How to Report t-Test Results: 5 Steps
Step 1: Name the test and the variables
Start by telling the reader which t-test you ran and what you compared. There are three types: independent-samples (two different groups), paired-samples (same people measured twice), and one-sample (one group against a known value). Name yours explicitly — readers should not have to guess.
Example: “An independent-samples t-test was conducted to compare exam scores between students who used flashcards and students who re-read their notes.”
Step 2: Report the descriptive statistics
Give the mean and standard deviation for each group (or each measurement), plus the sample sizes. Readers need this context before the test statistic makes any sense.
Example: “The flashcard group (n = 20) scored M = 78.50, SD = 8.20, while the re-reading group (n = 18) scored M = 71.30, SD = 9.10.”
Step 3: Write the core statistics sentence
Now report the test statistic, degrees of freedom, and p-value using the formula from above:
Example: “The difference was statistically significant, t(36) = 2.56, p = .015.”
For the independent-samples t-test, the degrees of freedom equal n1 + n2 − 2 (here 20 + 18 − 2 = 36). If your software reports adjusted degrees of freedom (for example after a Welch correction), report those instead — including decimals like t(31.40) = 2.49.
Step 4: Add the effect size and confidence interval
A p-value tells readers whether the difference is statistically significant; the effect size tells them whether it is practically meaningful. APA 7th edition strongly recommends reporting both, plus a confidence interval.
Example: “The effect was large, d = 0.83, 95% CI [1.51, 12.90].” (Cohen’s conventions: d ≈ 0.20 is small, 0.50 is medium, and 0.80 is large — but interpret against your field’s norms, not the labels alone.)
Step 5: Interpret the result in plain words
Close with one sentence a non-statistician can understand. Say what the numbers mean, not just what they are.
Example: “Students who used flashcards scored, on average, 7.20 points higher than students who re-read their notes.”
Worked Example 1: Independent-Samples t-Test
Let us put all five steps together into the paragraph you would actually paste into your results section:
“An independent-samples t-test was conducted to compare exam scores between students who used flashcards (n = 20, M = 78.50, SD = 8.20) and students who re-read their notes (n = 18, M = 71.30, SD = 9.10). Students in the flashcard group scored significantly higher than students in the re-reading group, t(36) = 2.56, p = .015, d = 0.83, 95% CI [1.51, 12.90]. On average, flashcard users scored 7.20 points higher, a large effect.”
That is the whole result — test named, descriptives given, statistics reported, effect size included, interpretation added. One paragraph, everything the reader needs.
Worked Example 2: Paired-Samples t-Test
Now imagine the same 20 students took a vocabulary quiz before and after a two-week flashcard program. Because the same people were measured twice, this needs a paired-samples t-test:
“A paired-samples t-test was conducted to compare vocabulary quiz scores before (M = 52.30, SD = 9.40) and after (M = 61.70, SD = 8.60) a two-week flashcard program (n = 20). Scores improved significantly after the program, t(19) = 4.12, p < .001, d = 0.92. On average, scores rose by 9.40 points.”
Note two things here. First, the degrees of freedom are n − 1 (20 − 1 = 19) because we are analyzing difference scores from the same participants. Second, the p-value was tiny, so we write p < .001 instead of copying SPSS’s p = .000 — more on that trap in the mistakes section.
Worked Example 3: One-Sample t-Test
Finally, suppose you want to know whether your sample of 50 university students differs from the national average IQ of 100. This calls for a one-sample t-test — and it is a good chance to show how to report a non-significant result, because those count too:
“A one-sample t-test was conducted to compare the sample’s IQ scores (n = 50, M = 103.40, SD = 12.80) against the national average of 100. The sample mean did not differ significantly from the national average, t(49) = 1.88, p = .066, d = 0.27.”
Notice the language: “did not differ significantly” — not “there was no difference” and not “the hypothesis was proven.” A non-significant result means the evidence was inconclusive, so keep your wording honest.
Quick Templates: Copy, Fill In, Done
Once you have done it a few times, reporting becomes automatic. Until then, use these templates:
| t-Test type | Template |
|---|---|
| Independent-samples | An independent-samples t-test showed that [Group A] (M = __, SD = __) scored significantly [higher/lower] than [Group B] (M = __, SD = __), t(__) = __, p = __, d = __. |
| Paired-samples | A paired-samples t-test showed that scores [increased/decreased] significantly from [Time 1] (M = __, SD = __) to [Time 2] (M = __, SD = __), t(__) = __, p = __, d = __. |
| One-sample | A one-sample t-test showed that the sample mean (M = __, SD = __) [did/did not] differ significantly from the population value of __, t(__) = __, p = __, d = __. |
6 Common Mistakes (and How to Fix Them)
- 1. Writing p = .000. SPSS displays three decimals, so tiny p-values show as .000. APA forbids this — write p < .001 instead. A p-value is never exactly zero.
- 2. Putting a leading zero on the p-value. Wrong: p = 0.015. Right: p = .015. Because a p-value can never exceed 1, APA drops the zero. (Keep the zero for means and SDs, which can exceed 1.)
- 3. Forgetting the degrees of freedom. Wrong: “t = 2.56, p = .015.” Right: “t(36) = 2.56, p = .015.” Without df, the t-value is uninterpretable — always include them in parentheses.
- 4. Italicizing everything. Italicize the symbols (t, p, M, SD, d) — not the numbers, parentheses, or the CI label. t(36) = 2.56 is right; t(36) = 2.56 is wrong.
- 5. Reporting only the p-value. “Significant, p = .015″ tells the reader nothing about how big the difference was. Always pair it with means, standard deviations, and an effect size.
- 6. Ignoring the assumptions check. If Levene’s test was significant (unequal variances), do not report the standard t row — report the “equal variances not assumed” row from SPSS, which uses adjusted degrees of freedom, and mention that you did so.
FAQ
Should I report exact p-values or just p < .05?
APA 7th edition prefers exact p-values to two or three decimals (e.g., p = .015). Only use p < .001 for values smaller than .001, and keep the older threshold style (p < .05) only if a journal specifically asks for it.
Where in my thesis does the t-test go?
In the Results chapter (or Results section of a paper), right after you present the descriptive statistics. Save the meaning of the result for the Discussion — the Results section just reports what you found.
Do I need to report the effect size for every t-test?
APA 7th edition strongly recommends it, and most supervisors now expect it. Cohen’s d takes seconds to compute from your descriptives and tells readers whether your significant result actually matters in practice.
What if my result is not significant?
Report it exactly the same way — test, df, statistic, exact p-value, effect size. Non-significant results are findings too, and hiding them is a form of reporting bias.
Can I write “highly significant” for very small p-values?
Better not to. APA style avoids “highly significant” — just report the exact value (p < .001) and let the number speak for itself.
Key Takeaways
- The core sentence is always t(df) = value, p = value — with italicized symbols, no leading zero on p, and exact p-values to 2–3 decimals.
- Surround it with context: name the test, give means and SDs, add Cohen’s d and a confidence interval, and finish with one plain-language sentence.
- Never write p = .000, never drop the degrees of freedom, and never skip the effect size.
Master this one pattern and you can report any t-test — independent, paired, or one-sample — with confidence. And if you are choosing which test to run in the first place, our guide on how to choose the right statistical test walks you through that decision step by step.