How to report a linear regression in APA 7

What must appear when you report a multiple regression, what the table looks like, and a full example you can adapt to your data.

Part of the series: Statistical Analysis for a Seminar Paper or Thesis: The Complete Guide

Reporting a regression has two parts: the model as a whole (how much it explains and whether it's significant) and each predictor on its own (its unique contribution).

What to take from the SPSS output

  • From the Model Summary table: R² and Adjusted R Square.
  • From the ANOVA table: F, the degrees of freedom (Regression and Residual) and the significance.
  • From the Coefficients table: for each predictor, B, the standard error (Std. Error), β, t and the significance. It's worth also requesting 95% confidence intervals for B, under Statistics → Confidence intervals.

Reporting the model

The regression model was significant and explained 24% of the variance in burnout, R² = .24, F(3, 146) = 15.37, p < .001.

The degrees of freedom of F are the number of predictors (here 3) and N minus the number of predictors minus 1 (here 150 − 3 − 1 = 146).

Reporting a single predictor

Workload significantly and positively predicted burnout, β = .32, t(146) = 4.12, p < .001, such that higher workload was associated with higher burnout.

The degrees of freedom of t are the same as the residual degrees of freedom of the model.

The table

With more than two or three predictors, present a table and refer to it in the text. An APA 7 table:

Predictor B SE β t p 95% CI for B
Workload 0.41 0.10 .32 4.12 < .001 [0.21, 0.61]
Social support −0.28 0.09 −.24 −3.07 .003 [−0.46, −0.10]
Seniority 0.02 0.03 .05 0.66 .510 [−0.04, 0.08]
  • The table number in bold on one line, and the title in italics on the line below.
  • No vertical lines. Horizontal lines only above and below the column headings and at the end of the table.
  • A note below the table if abbreviations need explaining.

Writing rules that apply to every report

  • Statistical symbols in italics: B, β, t, F, p, R².
  • A zero before the decimal point only for values that can exceed 1. So B = 0.41, but β = .32 and p = .003.
  • Two decimal places for most values, three for p.
  • Before reporting, check the model's assumptions: linearity, normality of residuals, equal variance and multicollinearity (VIF). If any of them doesn't hold, say so.

Next steps

For correlation vs regression and running it in SPSS: correlation or regression?. A dichotomous outcome? Logistic regression. With a mediator or moderator: PROCESS. And formatting the table and placing it in the chapter: APA tables and figures and how to write a results chapter.

Stuck? An expert can go through it with you.

Tell us what you need, and we'll come back with an expert in Statistical analysis and a quote.

Send a request

Last updated: