PN 3002 EXAM 6 2026 STUDY GUIDE SOLVED
QUESTIONS AND CORRECT ANSWERS
GRADED A+
◉ Nonparametric Stats Covered in 3001/3002 (& equivalences).
Answer: - Wilcoxon rank-sum & Mann-Whitney U test-Equivalent to
independent t test
- Kruskal-Wallis test- Equivalent to independent one-way ANOVA
- Wilcoxon signed-rank test- Equivalent to paired t test
- Friedman's test- Equivalent to repeated-measures one-way ANOVA
- Spearman's r
- Kendall's tau
◉ tied ranks.
Answer: If two scores are identical, they should be assigned tied
ranks
Ex: If 11th and 12th largest scores are the same, both receive a rank
of 11.5
◉ Mann-Whitney U & Wilcoxon Rank-Sum Test process.
Answer: 1- Scores are ranked across both groups
2- Ranks are summed for each of the groups
,3- Sum of ranks for one group is compared with the sum of ranks
that would be expected, if the two groups had been identical
4- The difference between the observed sum of ranks and the
expected sum of ranks is converted to a z score by dividing it by the
standard error
5- Having determined the z score, significance can be determined
using the normal distribution
◉ Kruskal-Wallis Test prcoess.
Answer: 1- Rank all scores across all groups
2- Sum the ranks within each group
3- Sum of ranks are compared (gives test statistic = H)
4- use the chi-squared distribution with k-1 degrees of freedom to
generate a p value
◉ Wilcoxon Signed-Rank Test process.
Answer: 1- For each individual, calculate the difference in score
between cond. A & B
2- Next, rank the absolute differences across all scores
3- Finally, sum the absolute ranks of the scores that increased, and
the ranks of the scores that decreased
4- Find the test statistic, which is the sum of the positive ranks (T+)
5- To determine significance, we calculate the mean sum of ranks
(mean of T− and T+) and the standard error (SET ̄)
, 6- These sum of ranks are converted to a z score using the following
formula
7- Wilcoxon signed-rank test generates a z score, so can be
converted to a probability using the normal distribution
◉ Friedman's ANOVA process.
Answer: 1- For each individual, scores are ranked across all
treatment conditions
2- Ranks are summed for each treatment condition
3- The sums of ranks for the different treatment conditions are used
in a formula that generates a test statistic (Fr)
4- To generate a p value, use the chi-squared distribution with k-1
degrees of freedom
◉ χ2 test statistic combines two aspects of the data.
Answer: 1- Number of observations that differ from what is
expected, based on the null hypothesis
2- Proportion of the observations that differ from what is expected,
based on the null hypothesis
◉ We will apply χ2 test statistics to two experimental designs
(include df info).
Answer: (1) No predictor variable, categorical outcome variable
- Degrees of freedom = k - 1
QUESTIONS AND CORRECT ANSWERS
GRADED A+
◉ Nonparametric Stats Covered in 3001/3002 (& equivalences).
Answer: - Wilcoxon rank-sum & Mann-Whitney U test-Equivalent to
independent t test
- Kruskal-Wallis test- Equivalent to independent one-way ANOVA
- Wilcoxon signed-rank test- Equivalent to paired t test
- Friedman's test- Equivalent to repeated-measures one-way ANOVA
- Spearman's r
- Kendall's tau
◉ tied ranks.
Answer: If two scores are identical, they should be assigned tied
ranks
Ex: If 11th and 12th largest scores are the same, both receive a rank
of 11.5
◉ Mann-Whitney U & Wilcoxon Rank-Sum Test process.
Answer: 1- Scores are ranked across both groups
2- Ranks are summed for each of the groups
,3- Sum of ranks for one group is compared with the sum of ranks
that would be expected, if the two groups had been identical
4- The difference between the observed sum of ranks and the
expected sum of ranks is converted to a z score by dividing it by the
standard error
5- Having determined the z score, significance can be determined
using the normal distribution
◉ Kruskal-Wallis Test prcoess.
Answer: 1- Rank all scores across all groups
2- Sum the ranks within each group
3- Sum of ranks are compared (gives test statistic = H)
4- use the chi-squared distribution with k-1 degrees of freedom to
generate a p value
◉ Wilcoxon Signed-Rank Test process.
Answer: 1- For each individual, calculate the difference in score
between cond. A & B
2- Next, rank the absolute differences across all scores
3- Finally, sum the absolute ranks of the scores that increased, and
the ranks of the scores that decreased
4- Find the test statistic, which is the sum of the positive ranks (T+)
5- To determine significance, we calculate the mean sum of ranks
(mean of T− and T+) and the standard error (SET ̄)
, 6- These sum of ranks are converted to a z score using the following
formula
7- Wilcoxon signed-rank test generates a z score, so can be
converted to a probability using the normal distribution
◉ Friedman's ANOVA process.
Answer: 1- For each individual, scores are ranked across all
treatment conditions
2- Ranks are summed for each treatment condition
3- The sums of ranks for the different treatment conditions are used
in a formula that generates a test statistic (Fr)
4- To generate a p value, use the chi-squared distribution with k-1
degrees of freedom
◉ χ2 test statistic combines two aspects of the data.
Answer: 1- Number of observations that differ from what is
expected, based on the null hypothesis
2- Proportion of the observations that differ from what is expected,
based on the null hypothesis
◉ We will apply χ2 test statistics to two experimental designs
(include df info).
Answer: (1) No predictor variable, categorical outcome variable
- Degrees of freedom = k - 1
