BST 571 Exam 2 study set Questions with Complete Solutions (Latest 2025) Which TWO of the following would be nominal variables?
a) Weight in kilograms b) Class rank c) Blood type d) Time
from start of trial till a criterion is attained e) Kind of medical insurance the subject has - Correct Answers ✅Blood type and Kind of medical insurance the subject has A researcher wants to compare children who are exposed to domestic violence to those who are not, on the percentage in each group who have to repeat first grade. The researcher
should use:
- A contingency table Chi-square
- An unpaired t-test
- A paired t-test
- A Mann-Whitney U test
- A Wilcoxon Signed Rank test
- Median test - Correct Answers ✅A contingency table Chi-
- A contingency table Chi-square
- An unpaired t-test
- A paired t-test 1 / 4
square A researcher is comparing children and adults (a group of each) on the mean number of pieces of candy each eats during a local holiday. In both groups, many eat a few pieces of candy, a smaller number eat 6-10 pieces, and a few eat a lot of candy (up to 43 pieces).
BST 571 Exam 2 study set Questions with Complete Solutions (Latest 2025)
- A Mann-Whitney U test
- A Wilcoxon Signed Rank test
- Median test - Correct Answers ✅A Mann-Whitney U test
A researcher measures two variables, A and B, for a group of subjects under condition 1 and condition 2 in a crossover design. If variable A had Gaussian distributed scores (and differences) while variable B did not (and could not easily be
transformed to Gaussian) then: What procedure to see if the
average A was higher in condition 1 than condition 2? How about variable B?
- A contingency table Chi-square
- An unpaired t-test
- A paired t-test
- A Mann-Whitney U test
- A Wilcoxon Signed Rank test
- Median test - Correct Answers ✅A paired t-test
A Wilcoxon Signed Rank test A researcher has two groups of gerbils. The dependent variable is the number of times each animal has bitten a
graduate student. The data is:
Group A: 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 2, 2 2 / 4
BST 571 Exam 2 study set Questions with Complete Solutions (Latest 2025)
Group B: 0, 0, 0, 0, 0, 0,0, 0, 0, 0, 0, 0, 1
This pattern is best suited to which test to compare the groups?
- A contingency table Chi-square
- An unpaired t-test
- A paired t-test
- A Mann-Whitney U test
- A Wilcoxon Signed Rank test
- Median test - Correct Answers ✅Median Test
When the data are badly skewed, but still basically numeric, and the sample size is too small to "trust robustness" the
best choice is:
- A parametric test, like a t-test.
- A rank test, like the Mann-Whitney U.
- A counting test, like the Median test. - Correct Answers
✅A rank test like Mann Witney U.We expected that 20% of newly diagnosed cancers of type X will be stage I, 50% stage II and 30% stage 3. You collect some data to investigate this hypothesis. What statistical test would be appropriate? - Correct Answers ✅Goodness of fit Chi-square. 3 / 4
BST 571 Exam 2 study set Questions with Complete Solutions (Latest 2025) Of the 20 subjects, 10 changed positively, 1 negatively, and 9
didn't change, p < 0.01. What test is suggested? (Hint: no
quantification of the magnitude of change is available, just these numbers). - Correct Answers ✅McNemar's test "100 subjects had a certain mutation; another 300 did not.30% of those with the mutation exhibited biochemical response X, as compared to 20% of non-mutation subjects, p < 0.01 by Contingency table Chi-square." The best
interpretation is:
- There were significantly more un-mutated than mutated
- More than 10% of the non-mutated group had response X.
- There are no significant differences.
- An error was made -- data involving percents should be
- An error was made. This should have been a Goodness of
- A greater proportion of those with the mutation exhibited X
subjects.
analyzed with t-tests, not Chi-square.
Fit Chi-square.
than those without the mutation. - Correct Answers ✅A greater proportion of those with the mutation exhibited X than those without the mutation.
A typical null hypothesis in a Goodness of Fit Chi-square says:
- There is no mean difference between the groups.
- There is no difference in percentages succeeding at a task
- / 4
between groups.