BIOL 2600 Lecture Notes - Null Hypothesis, John Tukey, Analysis Of Variance
Document Summary
Anova is similar to a t-test, the only difference is that anova is used when there are more than two groups. Anova = global test doesn"t tell you which group is diff just that one is. To determine what group is different use post-hoc test. Null hypothesis: no difference between the mean/average of the groups. Look at the tukey pairwise test to identify the differences. Use regression when : x and y are clear variables, x is controlled throughout the experiment (x is a clear variable) X is independent and y depends on what the x variable is. Goal of regression = determine how much dependence/the pattern of y can be determined by the influence of x. Basically finding the coefficient of determination (ranges from 0 - 1) 0 = 0% of y"s pattern is explained by x. 1 = 100% of y"s pattern is explained by x.