Quiz #7
Answers are given in the answer boxes below.
Scroll down for explantions for the answers (given in red).
Be sure to put your name on this quiz. Write the letters corresponding
to the correct answers in the answer boxes below.
| |
answer |
| Question 1 |
B |
| Question 2 |
B |
Question 1
Indicate which of the following two statements is true (only one of
them is true) by placing the letter
A or
B
in the proper answer box above.
- [ A ] A test is conservative in a specific situation if the
actual size of the test (the maximum probability of making a type I
error if the null hypothesis is true) is greater
than the nominal level of
the test (which is the stated allowance for the maximum probability of a type I
error).
- [ B ] A test is conservative in a specific situation if the
actual size of the test (the maximum probability of making a type I
error if the null hypothesis is true) is less than the nominal level of
the test (which is the stated allowance for the maximum probability of a type I
error).
If the actual probability of a type I error can be greater than the
nominal level of the test, the test can result in a type I error with
too large a probability, which gives us that the test is
anticonservative. A conservative test is a test for which the maximum
type I error rate is less than the nominal level of the test.
So a conservative test is a valid (but conservative) test, meaning that
it does not produce false rejections with too great of a probability,
but a conservative test can have poor power characteristics compared to
a valid test which is not conservative, or not as conservative.
Question 2
Indicate which of the following two statements is true (only one of
them is true) by placing the letter
A or
B
in the proper answer box above.
- [ A ] If a test is anticonservative in a specific situation, the
resulting p-value can be misleadingly large.
- [ B ] If a test is anticonservative in a specific situation, the
resulting p-value can be misleadingly small.
An anticonservative test is one which rejects the null hypothesis
with too great a probability when the null hypothesis is true. So it
produces small p-values with too great a probability --- so the p-values
are misleadingly small
Note: For each question, the two statements given differ by only
one word. (The words that differ are in italics.)