Skip to content
K Knidox Search…
Statistics · Reporting

APA Results Formatter

Format a t, F, χ², or r result in APA 7 style, ready to paste into your write-up.

Test
The value your software reported.
Exact p from your output. Anything below .001 is reported as p < .001.
For example an effect size: d = 0.62
APA 7 formatted result
t(24) = 2.31, p = .030

Statistics take two decimals and keep their leading zero; p and r drop it, because neither can exceed 1. Set the statistic symbol in italics when you paste this into your document.

This formats a result you already have — it does not compute or check it. Enter the exact p-value your software reported rather than a rounded one, so the rounding happens once.

APA 7 reports a test statistic with its degrees of freedom in parentheses, then the exact p-value: t(24) = 2.31, p = .030. Statistics keep their leading zero; p and r drop it, because neither can exceed 1. Anything below .001 is written p < .001.

The rules that trip people up

APA’s statistical formatting is not arbitrary. The leading zero is dropped for any quantity that cannot exceed 1 — p-values, correlations, proportions, alpha, R² — and kept for anything that can, such as t, F, and χ². So p = .030 but t = 2.31, and r = .42 but M = 0.42.

Degrees of freedom go in parentheses immediately after the symbol. An F test carries two: between-groups then within-groups, as F(2, 45). Chi-square is the exception that also carries the sample size: χ²(1, N = 120) = 5.43.

Exact p-values

APA 7 asks for exact p-values to two or three decimal places rather than the older p < .05 convention. The single exception is very small values, reported as p < .001 because further decimals convey nothing meaningful. Never write p = .000, which appears in SPSS output but asserts an impossibility — no p-value is exactly zero.

Worked example

Turning raw output into a sentence:

  1. 1
    Identify the test and its statistic. An independent-samples t-test returned t = 2.3147.
  2. 2
    Round the statistic to two decimals. 2.3147 becomes 2.31, keeping the leading zero if the value were below 1.
  3. 3
    Put the degrees of freedom in parentheses. With 24 degrees of freedom that gives t(24) = 2.31.
  4. 4
    Add the exact p-value without its leading zero. A p of 0.0298 is written p = .030 — two or three decimals, no leading zero.
  5. 5
    Add an effect size. APA expects one wherever possible: t(24) = 2.31, p = .030, d = 0.92.
  6. 6
    Italicise the symbols in your document. The statistic symbols and p are set in italics; the numerals and parentheses are not.

APA 7 formats for common tests

Symbols shown in plain text here; set them in italics in your manuscript. Greek letters such as χ are not italicised.

TestFormatExample
t-testt(df) = value, p = valuet(24) = 2.31, p = .030
ANOVAF(df₁, df₂) = value, p = valueF(2, 45) = 4.17, p = .022
Chi-squareχ²(df, N = n) = value, p = valueχ²(1, N = 120) = 5.43, p = .020
Correlationr(df) = value, p = valuer(38) = .42, p = .008
Very small pReported as an inequalityt(58) = 6.02, p < .001
With effect sizeStatistic, p, then the effect sizeF(2, 45) = 4.17, p = .022, η² = .16

What the formatter does not do

This tool formats a result you already have — it does not compute or verify it. Enter the exact p-value your software reported rather than a pre-rounded one, so rounding happens once rather than twice.

Two conventions worth carrying beyond the numbers themselves. Degrees of freedom for a correlation are N − 2, so a sample of 40 gives r(38); people often write the sample size there by mistake. And a result is reported in prose, not as a bare string: “Participants in the training condition scored higher than controls, t(24) = 2.31, p = .030, d = 0.92” communicates the finding, where the statistics alone do not.

Finally, report an effect size wherever one exists. APA 7 treats this as expected rather than optional, because a p-value alone says whether an effect is detectable, not whether it matters.

When do I drop the leading zero?
For any statistic that cannot exceed 1 — p-values, correlations, proportions, alpha, R². Values that can exceed 1, such as t, F, and χ², keep it. So p = .030 but t = 2.31.
Should I write p = .000?
Never. SPSS displays it when the value rounds to zero at three decimals, but no p-value is exactly zero. Write p < .001 instead.
How many decimals should a p-value have?
Two or three, and APA 7 asks for the exact value rather than a threshold. The only exception is very small values, which are reported as p < .001.
What are the degrees of freedom for a correlation?
N − 2, so a sample of 40 gives r(38). Writing the sample size in the parentheses instead is a common slip.
Why does chi-square include the sample size?
APA requires it because the chi-square statistic scales with N, so the same value means different things at different sample sizes. It is written χ²(1, N = 120) = 5.43.
What should be in italics?
The statistical symbols — t, F, r, p, M, SD — and not the numerals or parentheses. Greek letters such as χ, α, and η are not italicised.
Do I have to report an effect size?
APA 7 expects one wherever it can be computed. A p-value indicates whether an effect was detectable at your sample size; the effect size indicates whether it is large enough to matter.