01:830:200 Lecture Notes - Lecture 7: Type I And Type Ii Errors, Null Hypothesis, John Tukey
9 Oct 2018
School
Department
Course
Professor
Document Summary
Doing successive t-tests in a row for more than 2 means increases probability of making type i error (familywise error) Df b is number of levels of iv - 1. Df w is n - levels of iv. Null hypothesis is that the population means of all the groups are the same. Sum of squares within/between (look at 11-13 on pg. Lots of bullshit its pretty much you square all the scores in the group and subtract some shit. Post hoc [only if you reject the null] Tukey hsd: if any mean difference exceeds this number than we can conclude that it is statistically significant.
