What is statistical power in hypothesis testing?

What is statistical power in hypothesis testing?

• A. The probability of failing to reject the null
• B. The probability of correctly rejecting the null
• C. The probability of obtaining a significant result by chance
• D. The probability of Type II error

Answer: (B)

Statistical power is the probability that a test will reject the null hypothesis when there actually is a true effect. In practical terms, it measures how likely your study is to detect a real difference if one exists. Power is the complement of the Type II error rate (beta), so power = 1 - beta. Power increases when the true effect is larger, when you have a larger sample size, when there is less variability in the data, and when you set a higher significance level (alpha), though raising alpha comes with more false positives. In study planning, researchers often aim for around 0.8 (80%) power, meaning there’s an 80% chance of detecting the effect if it’s truly present. The other ideas are different: the probability of obtaining a significant result by chance is the Type I error rate (alpha); the probability of failing to reject the null is the Type II error rate (beta).

Statistical power is the probability that a test will reject the null hypothesis when there actually is a true effect. In practical terms, it measures how likely your study is to detect a real difference if one exists. Power is the complement of the Type II error rate (beta), so power = 1 - beta.

Power increases when the true effect is larger, when you have a larger sample size, when there is less variability in the data, and when you set a higher significance level (alpha), though raising alpha comes with more false positives. In study planning, researchers often aim for around 0.8 (80%) power, meaning there’s an 80% chance of detecting the effect if it’s truly present.

The other ideas are different: the probability of obtaining a significant result by chance is the Type I error rate (alpha); the probability of failing to reject the null is the Type II error rate (beta).