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.
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.
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.