How to Report Results in APA 7: Templates & Examples

To report results in APA 7 you write the same string every time: the symbol of the statistic in italics, the degrees of freedom in parentheses, the value to two decimals, the exact p value to three decimals and the effect size. Like this: t(58) = 3.42, p = .001, d = 0.89, or F(2, 87) = 5.23, p = .007, η²p = .11. If that string is complete, the result is properly reported; if the effect size is missing, a reviewer will ask you for it. Below you have the template for each analysis (t-test, ANOVA, correlation, regression, chi-square, non-parametric tests, EFA and SEM) so you can copy it and swap in your own numbers.

This is a skill any researcher who publishes needs to master, whether you work in psychology, health, biology, education or economics, and yet it is rarely taught systematically. You learn to run the analyses, you interpret SPSS or R output, but when the time comes to present those results in a manuscript, many find themselves lost among italics, decimal places, and typographic conventions that were never explained with sufficient clarity. The result is results sections riddled with formatting errors that, while they may seem minor, convey an image of carelessness and can lead reviewers to question the author's methodological competence.

The seventh edition of the APA Publication Manual, published in 2019, introduced some changes from previous editions, although the fundamental principles remain the same. The goal of these standards is to standardize the communication of results so that any reader, regardless of their institution or country, can interpret the data in the same way. Knowing these conventions not only improves the formal quality of our manuscripts but also facilitates the peer review process and increases the likelihood of acceptance in scientific journals.

General principles of statistical formatting

There are several cross-cutting rules that apply to virtually any statistical result reported in APA 7 format. Statistical symbols are always written in italics: t, F, p, r, d, η², M, SD, N, n. Numbers are generally rounded to two decimal places, with the exception of the p-value, which is reported to three decimal places (for example, p = .032). When the p-value is less than .001, it is simply written as p < .001, without attempting further precision. A convention that frequently causes confusion is the omission of the leading zero before the decimal point for statistics that cannot exceed a value of 1 (such as correlations, p-values, or bounded effect sizes). We write r = .45 and p = .012, not r = 0.45 or p = 0.012. However, statistics that can exceed 1 (such as means, standard deviations, or t and F values) retain the zero: M = 0.87, t(45) = 0.92.

Degrees of freedom are included in parentheses immediately after the statistical symbol, with no space: t(34), F(2, 87), χ²(3). In the case of the F statistic from ANOVA, two degrees of freedom are reported separated by a comma: the numerator (effect) and the denominator (error). Confidence intervals are expressed in brackets, preceded by the confidence level: 95% CI [2.34, 5.67]. And effect sizes must always accompany significance tests, something that APA 7 emphasizes with particular insistence.

How to report the most common analyses

For an independent samples t-test, the standard format includes the mean and standard deviation of each group, the t statistic with its degrees of freedom, the p-value, and the effect size (typically Cohen's d). A complete example would be: "The experimental group (M = 23.45, SD = 4.12) scored significantly higher than the control group (M = 19.87, SD = 3.98), t(58) = 3.42, p = .001, d = 0.89." Note that Cohen's d retains the leading zero because, although uncommon, it can theoretically exceed a value of 1. If you are unsure which values of d count as small or large in your field, see the guide on interpreting effect sizes, and for the long version with the SPSS output step by step, the one on reporting the t-test and chi-square in APA 7.

For a one-way ANOVA, the F statistic is reported with both degrees of freedom, the p-value, and the effect size (typically partial η² or generalized η²): "A significant effect of therapy type on symptom reduction was found, F(2, 87) = 5.23, p = .007, η²p = .11." When the ANOVA is significant and post hoc comparisons are performed, these must be reported indicating the method used (Bonferroni, Tukey, Games-Howell) and the results of each relevant comparison. It is not necessary to report all possible comparisons if only some are theoretically interesting, provided the selection is justified. The full write-up, including factorial and repeated measures designs, is in how to report an ANOVA in APA 7.

Correlations are typically presented in a correlation table when there are multiple correlations, with means and standard deviations included. For individual correlations mentioned in the text: "The relationship between trait anxiety and academic performance was negative and statistically significant, r(98) = -.34, p < .001." For multiple regressions, the unstandardized (B) and standardized (β) coefficients are reported, along with their standard errors, t-values, and p-values, generally in table format. The model R² and its overall significance through the F test must also be included. The table template and the wording that goes with it are in how to report a regression in APA 7.

Quick templates for chi-square, non-parametric tests and SEM

Chi-square test of independence. "The distribution of responses differed between groups, χ²(2, N = 180) = 9.87, p = .007, Cramer's V = .23." Inside the parentheses go the degrees of freedom and the sample size, separated by a comma, and the effect size (φ for 2 x 2 tables, Cramer's V for larger ones) is compulsory too. A very common shortcut is to give only the χ² and the p value and leave it there.

Non-parametric tests. "Group A scored higher (Mdn = 24.00) than group B (Mdn = 19.50), U = 312.50, z = -2.87, p = .004, r = .29." When the test is non-parametric, central tendency is described with the median, not the mean, and the effect size is derived from z. The detail for Mann-Whitney, Wilcoxon, Kruskal-Wallis and Friedman is in how to report non-parametric tests in APA 7.

Structural equation models. Here what you report is not a single statistic but model fit: "The model showed acceptable fit, χ²(84) = 152.31, p < .001, CFI = .96, TLI = .95, RMSEA = .054 [.041, .067], SRMR = .048." Do not report the model χ² alone: with large samples it comes out significant almost every time and tells you nothing useful. Alongside fit you need the standardized coefficients with their standard errors and confidence intervals.

How to report an Exploratory Factor Analysis (EFA) in APA 7

Exploratory factor analysis is one of the analyses that raises the most doubts when reporting, because it involves several elements that must appear together. Before extraction, report sampling adequacy with the KMO (Kaiser-Meyer-Olkin) measure and Bartlett's test of sphericity: "The KMO was .89 and Bartlett's test was significant, χ²(45) = 1234.50, p < .001, justifying factorization." Next, specify the extraction method (for example, principal axis factoring or maximum likelihood) and the rotation method (varimax if factors are independent; oblimin or promax if correlation between them is allowed), together with the criterion for deciding the number of factors (eigenvalues > 1, parallel analysis, or the scree plot).

The core of the report is a factor loadings table with items in rows and factors in columns, showing the loadings (loadings below .30 or .40 are usually suppressed for readability) and, where appropriate, the communalities. In the text, report the total variance explained and that of each factor: "The three-factor solution explained 58.4% of the total variance." Reporting EFA incompletely (omitting the KMO, not stating the rotation, or not justifying the number of factors) is one of the common reasons a reviewer asks you to redo the section. For the full step-by-step write-up (with the loadings table, KMO, explained variance and a worked example), follow the guide on how to report an EFA in APA 7. If you need to validate a questionnaire, this is exactly the kind of analysis I make reviewer-proof.

Tables and figures: essential complements

APA 7 standards specify that tables should have a clean format with horizontal lines only at the top, below the headers, and at the bottom. Vertical lines are not used. Each table has a bold number (Table 1, Table 2...) and an italicized title that concisely describes its content. Footnotes are used to explain abbreviations, indicate significance levels, or provide any necessary clarification.

Figures have gained importance in communicating results, and APA 7 offers more detailed guidelines than previous editions regarding their format. Each figure requires a bold number and an italicized title, just like tables. Axes must be clearly labeled, legends must be self-explanatory, and if error bars are included, it must be specified what they represent (standard deviation, standard error, confidence interval). A common error is including figures that duplicate information from tables. The APA recommends choosing the most appropriate format for each result: tables are preferable when numerical precision is important, while figures are more effective for showing patterns, trends, or interactions. In a meta-analysis, the standard figure is the forest plot, which summarizes each study's effect and the pooled effect in a single image.

Common errors and how to avoid them

The most common error, by far, is omitting effect sizes. Many manuscripts report p-values as if they were the only relevant indicator, ignoring the explicit APA 7 recommendation to always include measures of effect magnitude. Another frequent error is reporting results as "marginally significant" when the p-value falls between .05 and .10. APA 7 discourages this practice because it perpetuates a dichotomous interpretation of significance and suggests precision where there is none. It is preferable to report the exact p-value and let the reader draw their own conclusions.

It is also common to find inconsistencies in the use of decimal places, errors in the formatting of degrees of freedom, or the inclusion of Levene's test or the Shapiro-Wilk test without any context. Preliminary assumption analyses should be mentioned, but it is not necessary to exhaustively report every diagnostic test. It is sufficient to indicate that the relevant assumptions were verified, briefly describe the results, and detail the corrective measures adopted in case of violations (for example, the use of the Greenhouse-Geisser correction for sphericity or robust standard errors for heteroscedasticity).

A good practice worth mentioning is the use of specific templates or checklists for verifying APA format before submitting a manuscript. Checking that each statistic includes all required elements (italicized symbol, degrees of freedom, exact value, p-value, effect size) avoids tedious corrections during the review process and conveys professionalism. Ultimately, reporting correctly is not just a matter of editorial aesthetics but of scientific transparency: enabling other researchers to evaluate, interpret, and, if necessary, reanalyze our results.

Before you submit: if your results section is already written, run it through the AI Paper Reviewer. It is a free Reviewer 2 style pre-review that tells you what would come back at you (missing effect sizes, incomplete statistics, tables that disagree with the text) while you can still fix it, instead of finding out three months later in the editor's letter.

If you need your dissertation results to pass APA 7 format without minor incidents flagged by the committee, my statistical consulting for doctoral theses reviews the tables and the full results section before submission.