Written and maintained by CASRAI Editorial Board
Last updated
Reporting statistics in APA style requires following standardized rules established in the seventh edition of the Publication Manual of the American Psychological Association (APA 7th). These guidelines govern how quantitative findings—ranging from descriptive summaries to inferential test statistics, exact probability values, effect sizes, and confidence intervals—must be structured within empirical manuscripts. Adhering to these conventions ensures that research findings are communicated unambiguously across psychology, behavioral sciences, and quantitative disciplines.
Before preparing a manuscript’s results section, researchers should ensure they understand how to choose a statistical test appropriate for their research design and distributional assumptions. A foundational grasp of what constitutes a statistical test provides the theoretical basis for presenting test statistics, degrees of freedom, and sample parameters accurately.
Purpose of Statistics Reporting in APA Style
The primary purpose of standardizing statistical reporting in scholarly manuscripts is to maximize clarity, reproducibility, and precision. In empirical research, readers must be able to evaluate the statistical evidence directly from the text and tables without ambiguity. APA Style achieves this by providing a unified syntax for test statistics, establishing rigorous standards for reporting effect sizes and precision estimates, and minimizing visual clutter through standardized symbol conventions.
In practice, a reader should be able to see not only whether a result reached statistical significance but also how large the effect was and how precisely it was estimated. The APA Style numbers and statistics handout sets the rules for the symbols, spacing, and rounding used to report those values.
Core Structure and Formatting Rules
When presenting quantitative results in text, authors must follow five formatting rules from APA’s official numbers and statistics handout, covering variables, spacing, leading zeros, and decimal precision:
- Italicization of Latin Symbols: All Latin letters used as statistical symbols or algebraic variables must be italicized (e.g., M, SD, t, F, p, r, d, N, n, z).
- Non-Italic Greek Characters: Greek letters representing population parameters or statistical tests must be set in plain, standard roman typeface (e.g., χ², η², ηp², α, β, μ, σ). Never use italic formatting for Greek characters.
- Spacing Around Mathematical Operators: Put a space before and after a mathematical operator such as an equals, plus, minus, greater-than, or less-than sign (e.g., p < .05). For a negative value, put a space only before the minus sign, not after it (e.g., −8.25).
- The Leading Zero Rule (Bounded vs. Unbounded Statistics): Do not include a leading zero before the decimal point for statistics that cannot mathematically exceed 1.0 (such as proportions, correlations, and the level of statistical significance). Conversely, include a leading zero for statistics that can exceed 1.0 (such as means and standard deviations). For example, write p = .042 and r = .38, but M = 0.85 and t = 0.94.
- Decimal Precision: In general, report means and standard deviations for data measured on integer scales (such as surveys and questionnaires) to one decimal. Report other means and standard deviations, correlations, proportions, and inferential statistics (t, F, chi-square) to two decimals (e.g., M = 24.15, SD = 4.20). Report exact p values to two or three decimals (e.g., p = .006 or p = .03), and report values below .001 as p < .001, as covered in our guide on how to report p-values. These are general guidelines: round as much as possible while considering prospective use and statistical precision.
Reporting Formats for Common Statistical Tests
The table below shows common reporting patterns for standard analyses. Treat the effect-size and interval columns as typical practice for each test rather than a fixed APA template, and check the journal’s own instructions.
| Statistical Test | APA 7th Syntax Pattern | Key Formatting Elements |
|---|---|---|
| Independent-Samples t-Test | t(df) = value, p = .xxx, d = value, 95% CI [LL, UL] | Degrees of freedom in parentheses; Cohen’s d reported with leading zero; confidence interval enclosed in square brackets. |
| Paired-Samples t-Test | t(df) = value, p = .xxx, dz = value, 95% CI [LL, UL] | State the paired-sample effect size alongside mean difference and standard deviation of differences. |
| One-Way or Factorial ANOVA | F(dfbetween, dfwithin) = value, p = .xxx, ηp² = .xx | Two degrees of freedom separated by comma; partial eta squared reported without leading zero; Greek η non-italic. |
| Chi-Square Independence Test | χ²(df, N = total) = value, p = .xxx, V = .xx | Non-italic Greek χ; degrees of freedom and total sample size inside parentheses; Cramer’s V effect size. |
| Pearson Correlation | r(df) = .xx, p = .xxx, 95% CI [.xx, .xx] | Degrees of freedom (N − 2) in parentheses; correlation coefficient without leading zero. |
| Multiple Linear Regression | R² = .xx, F(dfreg, dfres) = value, p = .xxx; β = .xx, t(df) = value, p = .xxx | Report overall model fit first; report standardized regression coefficient β (non-italic Greek) and predictor significance. |
| Mann-Whitney U Test | U = value, z = value, p = .xxx, r = .xx | Non-parametric test statistic U and standardized test statistic z both italicized; rank biserial or effect size r. |
For interval estimates, see our companion resource on how to report confidence intervals.
Annotated Examples of Reporting Statistics in APA
Authors should integrate statistics into narrative sentences or present them parenthetically. The examples below demonstrate how to construct complete statistical statements according to APA 7th standards:
Independent-Samples t-Test Example
Narrative Presentation:
An independent-samples t-test revealed that participants in the cognitive training condition completed the task significantly faster (M = 14.25 s, SD = 2.10) than participants in the passive control condition (M = 17.80 s, SD = 2.45), t(58) = 5.98, p < .001, Cohen’s d = 1.55, 95% CI [1.02, 2.07].
Parenthetical Presentation:
Daily mindfulness practice produced a significant reduction in self-reported state anxiety scores compared with the waitlist condition, t(78) = 3.12, p = .002, d = 0.70, 95% CI [0.24, 1.15].
Factorial Analysis of Variance (ANOVA) Example
A 2 × 2 between-subjects ANOVA was conducted to assess the effects of feedback type and time limit on problem-solving accuracy. The main effect of feedback type was statistically significant, F(1, 116) = 12.45, p < .001, ηp² = .10, indicating that formative feedback resulted in higher scores than summative feedback. The main effect of time limit was not significant, F(1, 116) = 0.82, p = .367, ηp² = .01. Furthermore, the interaction between feedback type and time limit did not reach significance, F(1, 116) = 1.64, p = .203, ηp² = .01.
Multiple Linear Regression Example
Multiple linear regression was calculated to predict perceived job performance based on cognitive ability and conscientiousness. The overall regression model explained a significant proportion of variance, R² = .28, F(2, 147) = 28.58, p < .001. Both cognitive ability (β = .34, t(147) = 4.62, p < .001) and conscientiousness (β = .29, t(147) = 3.94, p < .001) emerged as significant individual positive predictors of job performance.
Common Mistakes in APA Format Statistics
Common formatting errors include:
- Reporting p = .000: Statistical software (such as SPSS or SAS) often truncates low probability values to .000 in output tables. In APA Style, writing p = .000 is incorrect because a probability is never strictly zero; authors must report p < .001 instead.
- Adding Leading Zeros to Bounded Metrics: Including a zero before values that cannot exceed 1.0 (such as writing p = 0.03 or r = 0.45) departs from APA’s rule on leading zeros for statistics that cannot exceed 1. Write p = .03 and r = .45.
- Italicizing Greek Characters or Non-Variable Terms: Applying italics to Greek letters (writing χ² instead of χ² or α instead of α).
- Repeating Statistics in Text and Table: APA’s handout says not to repeat statistics in both the text and a table or figure.
Manuscript Proofreading Checklist
Before submitting an empirical manuscript to a peer-reviewed journal, verify each item on this statistical reporting checklist:
- [ ] All Latin test letters (M, SD, t, F, p, r, d, z) are italicized.
- [ ] All Greek symbols (χ², ηp², α, β, μ, σ) are in standard roman (non-italic) type.
- [ ] A space appears before and after each mathematical operator.
- [ ] Statistics that cannot exceed 1.0 (p, r, α) omit the leading zero.
- [ ] Statistics that can exceed 1.0 (such as M and SD) include a leading zero when below 1.
- [ ] Degrees of freedom are clearly stated inside parentheses for all inferential tests.
- [ ] Exact p values are reported to two or three decimals, with values below .001 reported as p < .001.
- [ ] No statistic is repeated in both the text and a table or figure.
Related Reading
For how estimation and inference relate to predictive modeling, see data science vs. statistics; for the underlying discipline, see what is statistics.
Frequently Asked Questions
How do you report an exact p-value vs. p < .001 in APA format?
In APA 7th edition, report exact p-values to two or three decimal places (e.g., p = .006 or p = .03). When p is smaller than .001, report it as p < .001. Do not use a leading zero before the decimal point.
When should you use a leading zero in APA statistics reporting?
Include a leading zero before the decimal point only when a statistic can mathematically exceed 1.0 (such as means and standard deviations). Omit the leading zero when the statistic cannot be greater than 1 (such as proportions, correlations, and the level of statistical significance).
Do you italicize Greek letters in APA style?
No. APA says to use standard (nonitalic) type for Greek letters, while letters used as statistical symbols or algebraic variables (such as M, SD, t, F, and p) are italicized.
How do you format confidence intervals in APA 7th edition?
The usual form is the confidence level, "CI", and bracketed lower and upper limits, for example 95% CI [2.45, 5.12]. The full treatment is in our guide on how to report confidence intervals.
Should statistical symbols be defined in the text?
No. APA says not to define symbols or abbreviations that represent statistics, such as M, SD, F, t, df, p, N, and n. Other abbreviations, such as ANOVA and CI, should be defined.
Sources
- American Psychological Association. (2024). APA Style numbers and statistics guide (7th ed. handout, last updated September 11, 2024). https://apastyle.apa.org/instructional-aids/numbers-statistics-guide.pdf. Checked 2026-11-07. It cites Publication Manual Sections 6.36 (decimal places) and 6.40-6.45 (reporting statistics).
- American Psychological Association. (2020). Publication Manual of the American Psychological Association (7th ed.). Statistics guidance: https://apastyle.apa.org/style-grammar-guidelines/numbers/statistics. Checked 2026-11-07.
