CRIM 320 Lecture Notes - Lecture 11: Analysis Of Variance, List Of Statistical Packages
16 Apr 2018
School
Department
Course
Professor
Document Summary
Eg. what if we wanted to assess whether an individual"s self-perceived level of mastery differs by. How would you compare differences of mastery across catholics/protestants/members of the. Crim320 lecture 11: one-way analysis of variance. What we need is a single staisticalt test to assess differences between more than 2 groups. The test can e considered as an elaborated test. Anova compares 3 things: diff between means, diff in values within samples, diff in values across samples. Before conducting anova, it is important to compare the variances across groups. Each obs is diff from the grand mean by some amount. 2 sources of this diff: total variance, between group variance, within-group variance, iv, random/unexplained error. Anova wll not work if the variances are too diff. The heart of anova: the sum of squares. 3 types: total sum of squares (sstotal, between-group sum of squares (ssbetween explained, within-group sum of squares (sswithin unexplained)
