How to Report Regression Results in APA 7 Format (With Examples)

To report regression results in APA 7 format, state the overall model fit (R², F, df, and p), then report each predictor's unstandardized coefficient (B), standard error, standardized coefficient (β), and p value — ideally in a table, with the key findings described in the text. If you've just run a regression and you're staring at an SPSS output box wondering which numbers actually go in your results section, you're not alone — the output gives you far more than APA wants you to report.

Key Takeaways

• A complete APA 7 regression write-up reports R², adjusted R², F(df1, df2), and p for the overall model, plus B, SE, β, and p for each predictor.
• Italicise all statistical symbols (R², F, B, β, p, t) and drop the leading zero on values that cannot exceed 1 (write p = .003, not 0.003).
• Report 95% confidence intervals for coefficients where possible — APA 7 encourages CIs alongside, or instead of, exact p values.
• For regressions with more than two or three predictors, use a table and keep the text to the overall model and the standout predictors.
• Tools like StatRyx choose the right regression, run it, and generate the APA 7 table and write-up automatically, so the notation is correct the first time.

What numbers do I need to report for a regression in APA 7?

APA 7 wants two layers of information: how well the whole model fits, and how each predictor behaves inside it.

For the overall model, report:
- R² (the proportion of variance in the outcome explained by the predictors)
- Adjusted R² (penalised for the number of predictors)
- F with its two degrees of freedom: F(df1, df2)
- The p value for the model

For each predictor, report:
- B — the unstandardized coefficient (the change in the outcome per one-unit change in the predictor)
- SE — the standard error of B
- β — the standardized coefficient (lets you compare predictors on the same scale)
- t and p — the significance test for that predictor
- 95% CI for B where space allows

You do not report every line of SPSS output. The ANOVA summary table, the "variables entered" box, and residual diagnostics stay out of the results section — they live in your analysis, not your write-up.

How do I write a regression result in the text?

For a simple regression with one predictor, a single sentence carries the whole result. Here is the template:

A simple linear regression was conducted to predict [outcome] from [predictor]. The model was statistically significant, R² = ., F(df1, df2) = ., p = .___. [Predictor] significantly predicted [outcome], B = ., SE = ., β = ., p = .___.

The rules that trip people up:
- Italics: R, F, B, β, t, p are all italicised. The parentheses and numbers are not.
- Leading zeros: drop the 0 before the decimal for anything bounded by ±1 — so p = .021, β = .34, R² = .18. Keep the leading zero for F and B if they can exceed 1.
- Exact p values: report p to two or three decimals (p = .003). Only use p < .001 when the value is smaller than that.

Worked example: predicting exam scores from study hours and sleep

Suppose you surveyed 120 undergraduates and ran a multiple regression predicting final exam score from weekly study hours and average nightly sleep. Your output gives:

• R² = .34, adjusted R² = .33
• F(2, 117) = 30.14, p < .001
• Study hours: B = 1.82, SE = 0.29, β = .45, t = 6.28, p < .001, 95% CI [1.25, 2.39]
• Sleep: B = 2.10, SE = 0.88, β = .19, t = 2.39, p = .018, 95% CI [0.36, 3.84]

Here is how you'd write it in APA 7:

A multiple linear regression was conducted to predict final exam score from weekly study hours and average nightly sleep. The overall model was statistically significant and explained 34% of the variance in exam scores, R² = .34, adjusted R² = .33, F(2, 117) = 30.14, p < .001. Weekly study hours was a significant positive predictor, B = 1.82, 95% CI [1.25, 2.39], β = .45, p < .001, as was average nightly sleep, B = 2.10, 95% CI [0.36, 3.84], β = .19, p = .018.

What each number means in plain language: R² = .34 means study hours and sleep together account for a third of why exam scores differ between students. B = 1.82 for study hours means each additional weekly study hour predicts a 1.82-point increase in the exam score, holding sleep constant. β = .45 vs β = .19 tells you study hours matters more than sleep because its standardized coefficient is larger. And p < .001 means this pattern is very unlikely to be a fluke of your sample.

When should I use a table instead of text?

Use a table when you have three or more predictors, or when you're comparing nested models. APA 7 regression tables typically have one row per predictor and columns for B, SE, β, t, p, and the 95% CI. Keep the prose for the overall model fit and one or two headline predictors, then point readers to the table for the rest.

Here's a copy-ready APA 7 regression table for the example above:

Note. N = 120. R² = .34, adjusted R² = .33, F(2, 117) = 30.14, p < .001. CI = confidence interval.

The table title in a real manuscript sits above the table in italics (e.g. Table 1. Multiple Regression Predicting Final Exam Score), and the note sits below.

Simple vs multiple regression reporting: what changes?

The core difference: with multiple predictors you add adjusted R² (because R² inflates with every variable you add) and you lean on a table, since cramming four predictors into a sentence is unreadable.

How do I avoid the most common APA 7 mistakes?

The errors reviewers flag most often are: forgetting to drop the leading zero on p and β, writing p = .000 (which is impossible — report p < .001 instead), mixing up B and β, and omitting the degrees of freedom on F. Reporting R² without the F test is also incomplete — the effect size and the significance test belong together.

If you're hand-formatting output from SPSS or copying from R console text, these mistakes are easy to make and easy to miss. StatRyx run