Module_3_Flashcards_ANOVA_Non_Parametric_Tests

Question 1

What is ANOVA?

Answer 1

ANOVA (Analysis of Variance) is a statistical test used to compare the means of three or more groups.

Question 2

Why is ANOVA used instead of multiple t-tests?

Answer 2

ANOVA controls for Type I error that increases when performing multiple t-tests.

Question 3

When is a one-way independent-measures ANOVA used?

Answer 3

It is used when comparing the means of three or more independent groups on one factor.

Question 4

What are the assumptions of ANOVA?

Answer 4

ANOVA assumes normality, homogeneity of variances, and independent observations.

Question 5

What does the F-ratio represent in ANOVA?

Answer 5

The F-ratio compares between-group variance to within-group variance. A higher F-ratio suggests significant differences between groups.

Question 6

What does a significant p-value (<0.05) in ANOVA mean?

Answer 6

A significant p-value means there is a statistically significant difference between at least two group means.

Question 7

What is the effect size in ANOVA and why is it important?

Answer 7

Effect size (e.g., η²) shows the magnitude of the difference between groups. It is important to interpret practical significance, not just statistical significance.

Question 8

How do you handle violations of homogeneity of variance in ANOVA?

Answer 8

If the assumption of homogeneity is violated, you can use a Welch ANOVA or transform the data.

Question 9

What are post-hoc tests?

Answer 9

Post-hoc tests are conducted after a significant ANOVA to determine which specific group means are different.

Question 10

Why are post-hoc tests necessary after ANOVA?

Answer 10

Post-hoc tests control for Type I error when making multiple comparisons between group means.

Question 11

Give an example of a post-hoc test.

Answer 11

Examples include Tukey’s HSD, Bonferroni correction, and Scheffé test.

Question 12

How do you interpret post-hoc test results?

Answer 12

Look for pairwise comparisons with p-values less than 0.05 to identify which groups differ significantly.

Question 13

What are planned comparisons in ANOVA?

Answer 13

Planned comparisons test specific hypotheses developed before data collection about group differences.

Question 14

When would you use planned comparisons?

Answer 14

Use them when you have specific predictions about the differences between certain group means.

Question 15

How are contrast weights used in planned comparisons?

Answer 15

Contrast weights assign values to group means based on hypothesized relationships, focusing the comparison on specific groups.

Question 16

How do you interpret planned comparisons?

Answer 16

Interpret planned comparisons by looking at the contrast weights, F-ratio, p-value, and effect size.

Question 17

What is trend analysis in ANOVA?

Answer 17

Trend analysis examines patterns of change (e.g., linear or quadratic trends) across levels of the independent variable.

Question 18

Why would a researcher perform a trend analysis?

Answer 18

Trend analysis helps identify systematic increases or decreases in the dependent variable across different groups.

Question 19

When is trend analysis appropriate?

Answer 19

It is appropriate when you expect a specific pattern or trend across ordered levels of a variable (e.g., time, dosage).

Question 20

How do you interpret trend analysis results?

Answer 20

You interpret significant trends by looking for systematic patterns in the data (e.g., a linear or quadratic trend).

Question 21

What is the Mann-Whitney U test?

Answer 21

The Mann-Whitney U test is a non-parametric test used to compare two independent groups when the data are not normally distributed.

Question 22

When should the Mann-Whitney U test be used?

Answer 22

It should be used when data are ordinal or not normally distributed and the assumptions of the independent t-test are violated.

Question 23

How do you interpret the results of a Mann-Whitney U test?

Answer 23

If the p-value is less than 0.05, there is a significant difference in the ranks of the two groups.

Question 24

What is the Wilcoxon Signed-Ranks test?

Answer 24

The Wilcoxon Signed-Ranks test is a non-parametric test used to compare two related groups (e.g., pre-test vs post-test) when data are not normally distributed.

Question 25

When should the Wilcoxon Signed-Ranks test be used?

Answer 25

It should be used when comparing two related groups and the assumptions of the paired-samples t-test are violated.

Question 26

How do you interpret the results of the Wilcoxon Signed-Ranks test?

Answer 26

A significant p-value (<0.05) indicates a significant difference between the two paired samples.

Question 27

What is the Kruskal-Wallis test?

Answer 27

The Kruskal-Wallis test is a non-parametric test used to compare three or more independent groups when the assumptions of ANOVA are violated.

Question 28

When should the Kruskal-Wallis test be used?

Answer 28

It should be used when comparing three or more independent groups with ordinal or non-normally distributed data.

Question 29

How do you interpret the Kruskal-Wallis test results?

Answer 29

A significant p-value (<0.05) means that at least one group differs from the others. Post-hoc tests can then be performed.

Question 30

What are confidence intervals, and why are they important?

Answer 30

Confidence intervals provide a range of values within which the true population parameter is likely to fall. They show the precision of an estimate.

Question 31

How do sample size and confidence intervals relate?

Answer 31

Larger sample sizes lead to narrower confidence intervals, indicating more precise estimates of the population parameter.

Question 32

What is statistical power, and why is it important?

Answer 32

Statistical power is the probability of detecting a true effect. High power reduces the likelihood of a Type II error.

Question 33

How can researchers increase statistical power?

Answer 33

Researchers can increase power by increasing sample size, using stronger manipulations, or setting a higher alpha level.

Question 34

What is the relationship between sample size and statistical power?

Answer 34

Larger sample sizes increase statistical power, making it more likely to detect a significant effect if it exists.