What is the α-level in hypothesis testing?

• A. the probability of making a Type II error
• B. the probability of making a Type I error (usually 0.05)
• C. the p-value threshold for significance
• D. the power of the test

Answer: (B)

The α-level is the probability of making a Type I error. That means it’s the pre-specified threshold for declaring significance: if the p-value from your test is at or below α, you conclude there is an effect and reject the null hypothesis. By setting α at, for example, 0.05, you’re willing to accept up to a 5% chance of mistakenly rejecting a true null hypothesis (a false positive) over many repeated studies. It’s chosen before collecting data and acts as the cutoff for significance you compare your p-value to. Lowering α reduces the chance of a false positive but can also reduce the test’s power, increasing the chance of a Type II error.