When to use it
- •Your outcome is continuous and you have two or more predictors.
- •You want the unique contribution of each predictor, controlling for the rest.
- •You need effect sizes rather than just a yes/no significance verdict.
Assumptions to check first
- •Linearity between each predictor and the outcome.
- •Independent, homoscedastic, approximately normal residuals — check the residual plots, not just the normality of raw variables.
- •No severe multicollinearity (VIF is the usual diagnostic; values above 5–10 are commonly flagged).
What to report
- •The overall model: R², adjusted R², and the F test with both degrees of freedom.
- •For each predictor: unstandardised B with its standard error, standardised β, t, and p.
- •Confidence intervals for B — APA 7 strongly favours interval estimates over bare p values.
APA 7 example
The model explained a significant proportion of variance in job satisfaction, R² = .34, adjusted R² = .32, F(3, 196) = 33.41, p < .001. Autonomy was the strongest predictor, B = 0.41, SE = 0.07, β = .38, t = 5.86, p < .001, 95% CI [0.27, 0.55].
Numbers are illustrative — the Advisor generates this sentence with your own results.
The mistake reviewers catch
R² always rises when you add predictors, which is why adjusted R² exists. And "controlling for" is a statement about your model, not about the world — an omitted confounder is still omitted.
Run it on your own data
Paste or upload your dataset — nothing leaves your browser. The Analysis Advisor checks the assumptions above, runs linear regression, and drafts the APA Methods and Results text with your numbers in it.
Open the Analysis Advisor →Related analyses
Frequently asked questions
When should I use linear regression?
Which predictors explain my continuous outcome, and how much? Your outcome is continuous and you have two or more predictors.
What do I need to report for linear regression in APA 7?
The overall model: R², adjusted R², and the F test with both degrees of freedom. For each predictor: unstandardised B with its standard error, standardised β, t, and p. Confidence intervals for B — APA 7 strongly favours interval estimates over bare p values.
What is the most common mistake with linear regression?
R² always rises when you add predictors, which is why adjusted R² exists. And "controlling for" is a statement about your model, not about the world — an omitted confounder is still omitted.