vol.io
Log inStart for free

Simple Linear Regression Calculator

Fit a straight line to your data and get everything a statistics course or reviewer expects: the regression equation, a full coefficient table (estimate, standard error, t, p, 95% confidence interval for both slope and intercept), R² and adjusted R², the model F-test and the residual standard error — the same numbers R's summary(lm()) and confint() produce, verified to at least six significant digits. Paste two columns (x then y) straight from a spreadsheet. The calculator also checks the normality of residuals with Shapiro-Wilk automatically and flags problems in plain language — a diagnostic step most online regression tools skip entirely.

AI Report

Let AI interpret your results: a downloadable Word document in APA 7 / business report format.

Generate AI report (5 credits)Creating an account is free and includes free monthly credits.

Frequently asked questions

How do I report regression results in APA 7 format?

Report the model test and the coefficient, e.g.: "Study time significantly predicted exam score, b = 0.88, SE = 0.02, t(10) = 45.93, p < .001. The model explained 99.5% of the variance, R² = .995, F(1, 10) = 2109.73, p < .001." The AI Report generates the full APA regression table and paragraph — the part that is most tedious to format by hand.

What is the difference between R² and adjusted R²?

R² is the share of variance in y explained by x; it can only increase when predictors are added. Adjusted R² penalizes model complexity and is the fairer value to report when comparing models. For simple regression with one predictor the two are close.

How do I interpret the slope?

The slope b is the expected change in y when x increases by one unit. Its 95% confidence interval tells you the plausible range; if the interval excludes 0 (equivalently p < .05), the association is statistically significant.

What assumptions does simple linear regression make?

Linearity, independent observations, constant error variance (homoscedasticity) and normally distributed residuals. This tool tests residual normality automatically; for the others, inspect a residual plot — patterns or funnels indicate violations.