How to Report T-Test Results in APA Style (7th Edition)

Why Reporting T-Tests Correctly Matters

Running a t-test in SPSS takes about ten seconds. Writing it up properly is where most students lose marks. Examiners, supervisors, and journal reviewers all read your results section looking for the same thing: the test statistic, its degrees of freedom, the exact p-value, the descriptive statistics behind the test, a confidence interval, and an effect size — all formatted the way the APA Publication Manual (7th edition) requires.

This guide walks you through reporting all three t-tests — one-sample, independent-samples, and paired-samples — in APA 7th edition style. You will see exactly which numbers to pull from your SPSS output, how to format each one, and you get copy-paste sentence templates for each test. The examples below use illustrative numbers so you can see the format; just swap in your own results.

What APA Expects: The Six Pieces Every T-Test Report Needs

Before the formatting rules, get the content right. A complete t-test report gives the reader enough to judge the result without re-running your analysis:

1. The test statistic — the t value itself.
2. Degrees of freedom — in parentheses immediately after t.
3. The exact p-value — not just “significant” or “not significant.”
4. Descriptive statistics — group means and standard deviations so the reader sees what the test was comparing.
5. A confidence interval — usually the 95% CI for the mean difference.
6. An effect size — Cohen’s d for t-tests, so the reader knows whether the difference is practically meaningful.

If any of these is missing, a careful reader cannot fully evaluate your finding. Aim to include all six every time.

Finding the Numbers in SPSS Output

Here is where each value lives in the SPSS output tables, click by click.

One-sample t-test

Menu path: Analyze → Compare Means → One-Sample T Test. Move your variable into the Test Variable(s) box, enter the comparison value in the Test Value field, click OK. In the “One-Sample Test” table, read the t column, the df column, and the Sig. (2-tailed) column (this is your p-value). The Mean Difference column gives the difference between your sample mean and the test value. The “One-Sample Statistics” table above it gives your sample’s mean and standard deviation.

Independent-samples t-test

Menu path: Analyze → Compare Means → Independent-Samples T Test. Move the outcome variable into Test Variable(s) and the grouping variable into Grouping Variable, then click Define Groups to enter the two group codes. In the “Independent Samples Test” table, first check Levene’s Test for Equality of Variances: if its p-value is above .05, read the top row (“Equal variances assumed”); if below .05, read the bottom row (“Equal variances not assumed”) — note that the degrees of freedom in the bottom row are adjusted and usually not a whole number. Take t, df, Sig. (2-tailed), Mean Difference, and the 95% Confidence Interval of the Difference. Group means and standard deviations come from the “Group Statistics” table above.

Paired-samples t-test

Menu path: Analyze → Compare Means → Paired-Samples T Test. Move both variables into the Paired Variables box and click OK. In the “Paired Samples Test” table, read t, df, Sig. (2-tailed), Mean (this is the mean of the difference scores), and the 95% Confidence Interval of the Difference. The “Paired Samples Statistics” table gives the mean and standard deviation of each condition separately.

Paste-ready SPSS syntax (with effect sizes and CIs)

If you use syntax, this block runs all three tests with 95% confidence intervals and effect sizes. In SPSS version 27 and later, /ES DISPLAY(TRUE) adds Cohen’s d to the output.

* One-sample t-test against a test value of 100.
T-TEST
  /TESTVAL=100
  /MISSING=ANALYSIS
  /VARIABLES=score
  /ES DISPLAY(TRUE)
  /CRITERIA=CI(.95).

* Independent-samples t-test comparing group 1 vs group 2.
T-TEST GROUPS=group(1 2)
  /MISSING=ANALYSIS
  /VARIABLES=score
  /ES DISPLAY(TRUE)
  /CRITERIA=CI(.95).

* Paired-samples t-test comparing pre and post scores.
T-TEST PAIRS=pretest WITH posttest (PAIRED)
  /MISSING=ANALYSIS
  /ES DISPLAY(TRUE)
  /CRITERIA=CI(.95).

How to Report a One-Sample T-Test in APA Style

A one-sample t-test compares your sample mean against a known or hypothesized value. Report the sample mean and standard deviation, the test value, the t statistic with degrees of freedom, the exact p-value, the confidence interval, and the effect size. Here is the format, with an illustrative example:

Example: A one-sample t-test compared students’ exam scores (M = 78.40, SD = 9.15) against the national benchmark of 70. Scores were significantly higher than the benchmark, t(49) = 6.48, p < .001, 95% CI [5.79, 10.99], d = 0.92.

Notice the narrative flow: describe what was compared first, then give the statistics in a fixed order — descriptives, test statistic, p-value, confidence interval, effect size — and end with a plain-language interpretation.

How to Report an Independent-Samples T-Test in APA Style

An independent-samples t-test compares the means of two separate groups. Report each group’s mean and standard deviation, the t value with degrees of freedom, the p-value, the confidence interval of the mean difference, and the effect size. If Levene’s test was significant and you used the adjusted row, mention that the equal-variances assumption was violated. Example:

Example: Students who used spaced practice (M = 82.30, SD = 8.40) scored significantly higher than students who crammed (M = 74.10, SD = 10.25), t(78) = 3.91, p < .001, 95% CI [3.98, 12.42], d = 0.87.

If Levene’s test had been significant, you would add a sentence like: “Levene’s test indicated unequal variances, so degrees of freedom were adjusted from 78 to 71.42” — and then report the t, adjusted df, and p from the “equal variances not assumed” row.

How to Report a Paired-Samples T-Test in APA Style

A paired-samples t-test compares two measurements from the same people. Report the mean and standard deviation of each condition (or of the difference scores), the t statistic with degrees of freedom, the p-value, the confidence interval, and the effect size. Example:

Example: Anxiety scores were significantly lower after the intervention (M = 14.20, SD = 4.60) than before it (M = 19.80, SD = 5.10), t(39) = 5.63, p < .001, 95% CI [3.59, 7.61], d = 0.89.

APA Formatting Rules for Statistical Results

Getting the numbers right is half the job; formatting them the APA way is the other half. These rules apply to every statistic you report:

• Italicize statistical symbols. t, df, p, M, SD, d, r, F, and n are all italicized. The numbers themselves are not.
• Degrees of freedom go in parentheses immediately after the statistic, with no space: t(49), never t (49).
• Put spaces around the equals sign: t(49) = 6.48, not t(49)=6.48.
• Report exact p-values to two or three decimal places: p = .032, p = .487. Never write p = .000 — SPSS prints .000 when the value rounds below .001, but APA requires p < .001.
• Omit the leading zero for values that cannot exceed 1.0: p = .032, r = .45. Keep the leading zero for values that can exceed 1.0: t(49) = 6.48, M = 78.40, d = 0.92.
• Round consistently — two decimal places is the standard for t, means, standard deviations, and effect sizes.
• Confidence intervals use brackets: 95% CI [5.79, 10.99], with the lower bound first and a comma between bounds.
• Name the test and the alpha level the first time: “A one-sample t-test (α = .05) was conducted…” After that, the statistics speak for themselves.

Effect Sizes and Confidence Intervals: What to Add and Why

The p-value tells you whether a difference is statistically detectable; it says nothing about whether the difference matters. That is the effect size’s job. For t-tests, Cohen’s d is the standard. Commonly used benchmarks interpret d ≈ 0.20 as small, ≈ 0.50 as medium, and ≈ 0.80 as large — but treat these as rough guides, not verdicts. A “small” effect can be practically important in your field, and a “large” one can be trivial. Always interpret the effect size in the context of your research question, not the benchmark alone.

The confidence interval adds precision information the p-value hides: it shows the range of plausible values for the true difference. A narrow interval around a large effect is strong evidence; a wide interval means your estimate is imprecise even if the p-value is significant. Reporting both the CI and the effect size turns a bare “it was significant” into a result your reader can actually evaluate — which is exactly what supervisors and reviewers are looking for.

Copy-Paste Reporting Templates

Replace the bracketed placeholders with your own values:

• One-sample: A one-sample t-test compared [variable] (M = [mean], SD = [SD]) against [test value]. [Interpretation sentence], t([df]) = [t], p = [p], 95% CI [lower, upper], d = [d].
• Independent-samples: [Group A] (M = [mean], SD = [SD]) [scored higher/lower than / did not differ from] [Group B] (M = [mean], SD = [SD]), t([df]) = [t], p = [p], 95% CI [lower, upper], d = [d].
• Paired-samples: [Variable] was [significantly higher/lower / not significantly different] [after vs. before / in condition B vs. condition A] (M = [mean], SD = [SD]) than (M = [mean], SD = [SD]), t([df]) = [t], p = [p], 95% CI [lower, upper], d = [d].

Common Mistakes That Cost Marks

• Writing p = .000. SPSS means “less than .001,” not zero. Always write p < .001.
• Reporting only the p-value. “The result was significant (p < .05)” tells the reader almost nothing. Include the test statistic, df, descriptives, CI, and effect size.
• Forgetting Levene’s test. For independent-samples t-tests, always check Levene’s test first and report from the correct row. Supervisors notice when you skip this.
• Mixing up one-tailed and two-tailed p-values. SPSS reports two-tailed p-values by default; report those unless you preregistered a one-tailed test.
• Claiming “no difference” from a non-significant result. A non-significant t-test means you did not detect a difference — not that the groups are identical. Report the CI and effect size, which may show a meaningful effect your sample was too small to confirm.

Frequently Asked Questions

Do I need to report the p-value if it is not significant?

Yes — always report the exact p-value, significant or not (for example, p = .214). Selective reporting of only significant results is misleading, and reviewers expect the full picture.

Should I report a one-tailed or two-tailed p-value?

Two-tailed, unless you had a clear directional hypothesis stated before collecting data. SPSS gives you the two-tailed p-value by default, which is what APA style expects in most cases.

What if Levene’s test is significant in my independent-samples t-test?

Read the “Equal variances not assumed” row of the SPSS output table. Report its t, adjusted (non-integer) df, and p, and mention in one sentence that Levene’s test indicated unequal variances so degrees of freedom were adjusted.

How do I report a t-test in a table instead of text?

Put the descriptive statistics (M, SD, n) in the table body, then add the inferential statistics in a table note, e.g., “Note. t(78) = 3.91, p < .001, d = 0.87.” The same six pieces of information apply — tables organize them, they do not replace them.

Is Cohen’s d required, or is the p-value enough?

APA 7th edition expects effect sizes alongside p-values. A p-value tells you a difference was detected; the effect size tells you how big it is. Report both every time.

Need this done on your data? Get a free quote → /services/