Question

In respect of type-I error in the field of medical statistics, which one of the following is NOT correct?

• A. It is also called alpha error.
• B. It is often assigned a value of 0.05 in studies.
• C. It is equal to 1 minus the beta error.
• D. It is used to determine sample size.

Hint:

Remember that 1 minus beta is the definition of statistical power, not of alpha.

Answer 1

The Correct Option is C

Type-I error, also called alpha, is the risk of a false positive, rejecting a null hypothesis that is actually true. We check each statement against this idea.

1. Also called alpha error: Correct, this is simply the standard alternate name.
2. Often assigned a value of 0.05: Correct, a 5 percent alpha level is the most common convention used in medical research.
3. Equal to 1 minus the beta error: Incorrect. The quantity 1 minus beta is the power of a study, the ability to detect a real effect, not the definition of alpha. Alpha and beta are set separately.
4. Used to determine sample size: Correct, both alpha and beta are plugged into the sample size formula along with the expected effect size.

So the one statement that does not hold up is the claim linking alpha to 1 minus beta, since that relationship actually defines power.

Let's summarize:

• Alpha, type-I error: chance of a false positive, commonly set at 0.05.
• 1 minus beta: this is power, a completely different quantity from alpha.

The correct answer is that statement (c) is NOT correct.