Types of Significance Tests

Question 1

Significance tests can either be parametric or nonparametric.

Parametric tests are usually measured because they are normally distributed.

Name 2 types of parametric tests and what type of data they are for.

Answer 1

1. Student’s t-test - paired or unpaired.
2. Pearson’s Product-Moment Coefficient - correlation.
3. ANOVA - paired or unpaired from multiple samples.

Question 2

Significance tests can either be parametric or nonparametric.

Name 4 types of non-parametric tests and what type of data they are for.

Answer 2

1. Mann-Whitney U Test - unpaired data
2. Wilcoxon signed-rank test - compares two sets of observations on a single sample
3. Chi-Squared Test - used to compare propotions or percentages.
4. Spearman, Kendall rank - correlation.

Question 3

In the Student’s paired & unpaired t-tests, what does paired and unpaired stand for?

Answer 3

• Paired data refers to data obtained from a single group of patients. (e.g. measurements before and after an intervention.)
• Unpaired data refers to data obtained from two different groups of patients. Comparing the response to treatment in two different groups.

Question 4

What is an ANOVA test?

Answer 4

A statistical test to demonstrate statistically significant differences between the means of several groups.

Question 5

How does an ANOVA test differ from a student’s t-test?

Answer 5

It allows the comparison of more than just two means.

(Student t-test only compares 2 means)

Question 6

Is an ANOVA test parametric or non-parametric?

Answer 6

Parametric