Your browser is out of date

This browser version is too old for Quizara to display the page correctly. Update to the latest version and refresh to continue.

Hypothesis Testing of Regression Coefficients

Updated · 2027 Level I · Quantitative Methods · Module 10 · Lesson 5 · 6 min lesson

On this page

What this lesson covers

A regression coefficient t-test compares an estimated slope or intercept with a hypothesized population value, using the coefficient's standard error.

The lesson tests zero and nonzero coefficient values with the appropriate alternative hypothesis. It explains coefficient standard errors, degrees of freedom, p-values and the effect of changing significance levels.

How do you test a regression slope?

Divide the estimated slope minus its hypothesized value by the slope's standard error to obtain a t-statistic with n − 2 degrees of freedom.

Slope t-statistic = (estimated slope − hypothesized slope) / standard error of slope

Estimated slope
Hypothesized slope
Standard error of the estimated slope

Standard error of slope = standard error of estimate / sqrt[sum of squared X deviations]

Standard error of the estimate
Observed X and its sample mean

For example, assume the simple-regression inference conditions hold for the six-company sample, with estimated slope 1.5, standard error of the estimate 4 and sum of squared X deviations 112. The slope standard error is 4 / sqrt(112) ≈ 0.378; testing a zero slope gives t ≈ 1.5 / 0.378 = 3.97, exceeding the two-sided 5% critical value 2.776 with 4 degrees of freedom, so reject the null.

Review Module 10 formulas in Quizara →

Common mistakes

  • Treating rejection of a zero slope as proof that the slope exceeds a different hypothesized value draws a conclusion that the test does not establish.
  • Treating failure to reject as proof that the null is true; it only means the sample did not give enough evidence against it.

Try a practice question

Choose an answer, then check it to read the explanation for every option.

Practice question

A simple linear regression gives an estimated slope coefficient of 0.86 and a standard error of the slope coefficient of 0.15. The t-statistic for testing H0:b1=0.50H_0: b_1 = 0.50 against HA:b1≠0.50H_A: b_1 \neq 0.50 is closest to:

Continue studying in Quizara

In Quizara the lesson comes with more practice questions on the topic, explanations for every option and an AI tutor for your follow-up questions.

Written by the Quizara Editorial Team under our editorial standards.