Health Sciences 3801A/B Lecture Notes - Lecture 8: Bonferroni Correction, Null Hypothesis, Analysis Of Variance
Document Summary
It is inflated because each comparison has a 5% chance of incorrectly rejecting the null hypothesis when the null hypothesis is true. There is a 26. 5% chance of incorrectly rejecting the null hypothesis on at least one of those comparisons. This increases the experiment-wise alpha to match the target alpha. You compare each t-test to an alpha that is equal to the target alpha/number of comparisons. This maintains the experiment-wise alpha at 0. 05. A factorial anova has more than one independent variable. It must have two or more levels. The independent groups anova is the same is the independent groups" t-test. The sampling of any given participant within the data is not dependent upon the presence of any other participant within the data. This is a critical assumption that can derail a single-factor anova and make your results meaningless. The sampling of individuals is dependent on a characteristic that links them to other participants in convenience sampling.