ANOVA (and their nonparametric equivalents) are a comparative analysis group that allows researchers to compare three

ANOVA (and their nonparametric equivalents) are a comparative analysis group that allows researchers to compare three or more independent variables at a time without increasing the margin of statistical error. The biggest way I could see ANOVAS used are if a researcher is looking for possible disparities from a topic. For example, which ethnicity has the most incidence for hypertension within a geographical area or healthcare facility?

The diagnosis of having hypertension would be an independent variable with two levels (yes/no). The ethnicity groups would be used as our dependent variable with three or more variable levels (Caucasian, African American, Hispanic, Native America etc). In using the ANOVA test (or their non-parametric equivalents) researchers would be able to identify groups that are at risk for disparities and begin to work to intervene.

The effectiveness of the interventions could also be evaluated by ANOVA tests as well, as they can be used in pre-test/post-test study designs as well. ANOVA has been shown to be fairly robust (Kellar & Kline, 2013). Even if the variables do not rigidly adhere to the assumptions required for the test, the results may still be close to the truth (Kellar & Kline, 2013).

1. What is the difference between a t-test and an ANOVA?
2. What does “homogeneity of variance” mean in relation to conducting a one-way ANOVA? Why do we want a high p-value for Levene’s test?
3. What is the difference between a Kruskal-Wallis test, a Friedman’s test, and the McNemar’s test? Under what conditions would you use each of these tests? How are these tests all similar? What do they all have in common?