Type I and Type II errors

Home > Statistics Every Writer Should Know > The Stats Board > Discusssion

Type I and Type II errors
Message posted by Jason (via 129.89.251.122) on October 29, 2001 at 3:08 PM (ET)

Why might Type I error and Type II error occur?

READERS RESPOND:
(In chronological order. Most recent at the bottom.)

Re: Type I and Type II errors
Message posted by Tomi (via 154.32.143.110) on October 29, 2001 at 5:41 PM (ET)

The most important kind of decision making process in statistics is known as hypothesis testing. In this process the following terms are used:

Null hypothesis (H0): This is the original belief, from which one only moves if there is sufficient evidence.

Alternative hypothesis (H1): This is another belief, different to H0.

In hypothesis testing a test statistic is calculated and a decision made depending on the value of this test statistic. Examples of test statistic include z-scores, t-statistic, X squared statistic.

In making a decision, there is a probability that the decision is wrong. This probability is measured by the Type I and Type II errors.

There are four possible scenarios:

H0 is true and you decide to believe H0: correct decision.

H0 is true and you decide to believe H1: wrong decision. THe probability of this type of error is called the Type I error.

H1 is true and you decide to believe H1: correct decision.

H1 is true and you decide to believe H0: wrong decision. THe probability of this type of error is called the Type II error.

Your $5 contribution helps cover part the $500 annual cost of keeping this site online.