Why do t‑tests and p‑values measure whether a regression coefficient differs from zero

Answer First

A t‑test compares an estimated coefficient to its hypothesized value, scaled by its standard error. The p‑value tells you how likely it is to observe a coefficient this extreme if the null hypothesis were true. If the p‑value is small, you reject the null.

Problem Setup

Simple regression: \[ y = \beta_0 + \beta_1 x + u. \] Hypothesis test: \[ H_0: \beta_1 = 0, \] \[ H_1: \beta_1 \neq 0. \] t‑statistic: \[ t = \frac{\hat{\beta}_1 – 0}{SE(\hat{\beta}_1)}. \] Decision rule:

  • Large |t| → reject \(H_0\)
  • Small p‑value → reject \(H_0\)

Step-by-Step Explanation

1. Estimate the regression

Obtain \(\hat{\beta}_1\) and its standard error \(SE(\hat{\beta}_1)\).

2. Compute the t‑statistic

Divide the coefficient by its standard error. Example: If \(\hat{\beta}_1 = 2.0\) and \(SE = 0.5\), then \(t = 4.0\).

3. Find the p‑value

Use the t‑distribution with \(n – 2\) degrees of freedom. Large |t| → small p‑value.

4. Compare to significance level

Common choices: 10%, 5%, 1%. If p < α → reject the null.

5. Interpret the result

Rejecting the null means the coefficient is statistically different from zero. Failing to reject means there is not enough evidence to conclude an effect.

Intuition

The t‑test asks: “Is this coefficient big relative to its noise?” If the coefficient is large compared to its standard error, it is unlikely to be zero.

Common Exam Mistakes

  • Thinking a large coefficient automatically means significance.
  • Confusing p‑values with effect size.
  • Using the wrong degrees of freedom.
  • Interpreting ‘fail to reject’ as ‘accept the null.’

Final Summary

To compute a t‑test, divide the coefficient by its standard error. The p‑value measures how extreme the t‑statistic is under the null. Small p‑values indicate statistical significance.

Students choose Statistics tutoring at California Graduate Tutor because we turn complex topics like probability, regression, hypothesis testing, and advanced statistical methods into clear, step-by-step solutions. Our approach emphasizes true understanding, exam readiness, and confidence across graduate and undergraduate coursework, with sessions tailored to your specific class and professor. Call 510 398 0006 or email tutor@californiagraduatetutor.com to get started.