# Thread: type 1, type 2 error question

1. ## type 1, type 2 error question

The latest in the “You may be a redneck if . . . ” genre has produced a simplequestionnaire. If more than 80% of the questions are answered in accordance with redneck “philosophy” (e.g., “You may be a redneck if grade 4 was the best 5 years of your life”) the test declares the respondent a redneck, but otherwise assumes that variation in the percentage of answers scored as redneck (including scores over 80%) is to be expected in people not actually rednecks. From that perspective, describe with reference to the test

(a) what corresponds to the criterion of the test?
(b) what would be a Type I error for the test?
(c) what would constitute a Type II error for the test?
(d) what would correspond to the power of the test?

attempt:
b) Being declared a redneck when you are actually not.

c) Being declared that you are not a redneck when you actually are.

i don't know what a and d would be....

2. ## Re: type 1, type 2 error question

a) I am not certain what is meant by "what corresponds to the criterion of the test?", but my guess is a redneck scoring greater than 80% and a non-redneck scoring 80% or below correspond to the criterion of the test.

b) An (actual) redneck who scores 80% or lower.

c) A non-redneck who scores greater than 80%.

d) Power is given by 1 minus the probability of a Type II Error. So, everything that is NOT a Type II Error corresponds to the power of the test. In this case, a redneck who scores above 80%, or any non-redneck (regardless of outcome) correspond to the power of the test.

Please refer to my chart, if it helps.