APA REPORTING GUIDE
How to report regression results in APA style
An APA-style regression report has two parts: the overall model (F test with both degrees of freedom, p, R², and adjusted R²) and each predictor (B, standard error or 95% CI, β, t, and p). Multiple regression results usually also go in a coefficients table.Explore research statistics & data analysis →FORMULA OR CORE IDEA
F(k, N − k − 1) = F value, p = .XXX, R² = .XX; predictor: B = X.XX, β = .XX, t(df) = X.XX, p = .XXXWORKED EXAMPLE
A multiple linear regression was run to predict exam score from weekly study hours and hours of sleep. The model was statistically significant, F(2, 12) = 27.22, p < .001, R² = .82, adjusted R² = .79. Study hours significantly predicted exam score, B = 3.26, 95% CI [2.11, 4.41], β = .76, t(12) = 6.19, p < .001, as did sleep, B = 4.29, 95% CI [1.67, 6.91], β = .44, t(12) = 3.57, p = .004.QUICK REFERENCE
APA regression templates
Replace the numbers with your own output. Italicise statistical symbols such as F, p, R², B, t, and SE in your final document.| What you report | Template | Notes |
|---|---|---|
| Overall model | F(2, 12) = 27.22, p < .001, R² = .82, adjusted R² = .79 | df = (k, N − k − 1) |
| Significant predictor | B = 3.26, 95% CI [2.11, 4.41], β = .76, t(12) = 6.19, p < .001 | SE can replace the CI: B = 3.26, SE = 0.53 |
| Non-significant predictor | B = 0.76, β = .09, t(11) = 1.25, p = .237 | Still report it; do not just say "ns" |
| Simple regression | R² = .45, F(1, 48) = 39.27, p < .001; B = 2.10, t(48) = 6.27, p < .001 | For one predictor, t² = F |
| Hierarchical step | ΔR² = .08, ΔF(2, 95) = 6.11, p = .003 | Report the change for each added block |
| Regression equation | Score = 21.14 + 3.26 × Hours + 4.29 × Sleep | Optional; useful for prediction |
STEP BY STEP
State the type of regression and the variables
Say whether it was a simple or multiple linear regression (or hierarchical, logistic, etc.), name the outcome (dependent variable), and list the predictors.
Report the overall model F test
Write F(df regression, df residual). The regression df equals the number of predictors k, and the residual df equals N − k − 1. With 2 predictors and 15 participants that is F(2, 12).
Give R² and adjusted R²
R² is the proportion of variance in the outcome explained by the model. Adjusted R² corrects for the number of predictors and is the better figure when you compare models. Drop the leading zero (R² = .82).
Report each predictor
For every predictor give the unstandardised coefficient B (with its SE or 95% CI), the standardised coefficient β, t with the residual df, and the exact p value. Say whether the relationship is positive or negative.
Mention the assumption checks
Briefly state that linearity, independence of residuals (Durbin-Watson), homoscedasticity, normality of residuals, and multicollinearity (VIF) were checked, and note anything you had to address.
Put the coefficients in a table for multiple regression
APA tables list each predictor in a row with columns for B, SE, β, t, p, and the 95% CI, and a note under the table giving N, R², adjusted R², and the F test.
COMMON QUESTIONS
Frequently asked questions
Should I report B or beta (β) in APA style?
Report both when you can. B (unstandardised) gives the effect in the original units, and β (standardised) lets readers compare predictors measured on different scales. In a table, include columns for B, SE, and β.
What degrees of freedom do I use for the t test of a coefficient?
Use the residual degrees of freedom, N − k − 1, the same as the second df of the overall F test.
How do I report a regression in one sentence?
"Study hours significantly predicted exam score, B = 3.26, t(12) = 6.19, p < .001, and the model explained 82% of the variance, R² = .82, F(2, 12) = 27.22, p < .001."
Do I report the constant (intercept)?
Include it in the coefficients table and the regression equation, but you usually do not need to interpret or discuss it in the text unless zero is a meaningful value for all predictors.
How do I report a non-significant regression model?
Report it the same way with the exact p value, for example F(3, 46) = 1.42, p = .249, R² = .08, and state that the predictors did not significantly predict the outcome.