This preview shows pages 1–5. Sign up to view the full content.
Name: ________________KEY___________________
1
6 May 2011
IE330 Spring 2011
Final Exam
Part 1: Mandatory (Chapters 15 and 16)
No calculators, closed book, closed notes.
Do not tear off any pages.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document Name: ________________KEY___________________
2
CHAPTER 15  Multiple choice
1.
(4 points) Concerning handling values that are equal to the hypothesized value in
nonparametric tests, which of the following are true? (Select all that are true.)
a.
In a sign test, ties with the hypothesized median should be discarded.
b.
In a Wilcoxon signedrank test, ties with the hypothesized median should
be given the average of the ranks they would have if they were slightly
different.
c.
In a KruskalWallis test, ties with the hypothesized value are discarded.
d.
In a Wilcoxon ranksum test, ties with the hypothesized value are given
the average of the ranks they would have if they were slightly different.
e.
None of the above is true.
2.
(4 points) Which of the following are assumptions of the KruskalWallis test?
(Select all that are true.)
a.
The populations must come from a normal distribution.
b.
The populations must have equal variances.
c.
The populations must be continuous.
d.
The populations must have similar distributions.
e.
None of the above is an assumption of the KruskalWallis test.
3.
(4 points) Which of the following would be a case where a nonparametric test
should not
be used? (Select all that are true.)
a.
When the variance cannot be stabilized using transformations.
b.
When all the assumptions of a parametric test are met.
c.
When there are outliers that cannot be removed.
d.
When the populations are categorical and not continuous.
e.
None of the above are true.
4.
(2 points) I have a distribution that is highly skewed, with several (valid) outliers
out on one tail of the distribution.
What is the best test to use to determine if the
median of that distribution is equal to 5.2? (Choose the best answer.)
a.
ttest
b.
Sign test
c.
KruskalWallis test
d.
Wilcoxon signedrank test
e.
None of the above
5.
(4 points) I have collected paired data and intended to do a paired ttest on the
resulting data.
However, in inspecting the data I find that the assumptions of the
paired ttest have not been met.
Which of the following tests could I use instead?
(Choose all that could be used.)
a.
Contingency table
b.
Runs test
c.
Rank transformation followed by a paired ttest
d.
Wilcoxon signedrank test
e.
None of the above
Name: ________________KEY___________________
3
6.
(4 points) Which of the following are true regarding a rank transformation?
(Choose all that are true)
a.
It is equivalent to a simple linear regression.
b.
It replaces the
values
of the data with their
ranks
.
c.
A low pvalue indicates that the rank transformation is significant.
d.
It requires that the data come from a symmetric distribution.
e.
None of the above are true.
7.
(2 points) Which of the following is the best test to use instead of a two sample t
test when the assumptions for running a parametric test have
not
been met, but
the distributions have the same shape and spread?
(Choose the best answer.)
a.
Sign test
b.
Wilcoxon signedrank test
c.
Wilcoxon ranksum test
d.
KruskalWallis test
e.
None of the above
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document Name: ________________KEY___________________
4
This is the end of the preview. Sign up
to
access the rest of the document.
This document was uploaded on 12/14/2011.
 Spring '08

Click to edit the document details