# Nursing 3340A/B Lecture Notes - Lecture 10: Type I And Type Ii Errors, Analysis Of Variance, Null Hypothesis

## Document Summary

Data analysis 2step: 3 or 4 or more groups, similar to t tests. T-test: compares means of two groups, e. g. outcomes in drug treatment vs control (placebo) groups, e. g. outcomes comparing two drugs. Anova: compare means of multiple groups, e. g. outcomes for multiple treatments, h0: all treatment group means are equal. Anova: no treatment effect: normal distribution, the means are similar across- not the same, still in that critical region, can still reject the null hypotheses. Why do we need anova: with multiple t-tests , each t-test has a chance of type i error, the more tests, the higher the chance of error. Anova, like t-test : must use interval or ratio data or quasi-interval/ratio, in practice, ordinal data are used if the scales are symmetric. Independent or dependent samples: the groups being compared should have similar sds, use the appropriate group. Increases a likelihood of finding significant differences where they exist (power) in anova: example: