For teaching I need to explain the following diagram to undergraduate students. The diagram is taken from the 7th Edition of Howell's "Statistical Methods for Psychology". The students are familiar with how to calculate the Total SS, Between SS, and Within SS for a One-Way Independent groups design, and also for a One-Way Repeated Measures design, but don't know anything more than that.
What's an intuitive way to explain SS decomposition for a Mixed ANOVA with 1 B/S and 1 W/S variable?
$\begingroup$
$\endgroup$
5
-
$\begingroup$ My question when looking at the figure is: What is $I$? It seems to be a within-subject factor, but without somewhat more detail this could be difficult. $\endgroup$– HenrikMay 12, 2015 at 8:59
-
$\begingroup$ I refers to Intervals, which is a within-subjects term referring to the fact that subjects were measured more than once. In this example, according to p466-70 of the book (books.google.com.au/books/about/…), 24 subjects from 3 groups were measured 6 times. $\endgroup$– user1205901 - Слава УкраїніMay 12, 2015 at 9:09
-
$\begingroup$ Those six pages I mentioned can be viewed at imgur.com/a/W1coZ. $\endgroup$– user1205901 - Слава УкраїніMay 12, 2015 at 9:10
-
1$\begingroup$ It is not correct didactically to jump from one-between-group-factor ANOVA to mixed between-within 2 factor ANOVA. You first have to explain the transition from one-between-group-factor ANOVA to one-repeated-measures-factor ANOVA - how it redistributes SS and diminishes SSerror to test the factor. $\endgroup$– ttnphnsMay 12, 2015 at 10:10
-
$\begingroup$ @ttnphns that makes sense, and I'd aim to do that as an initial step. Perhaps I should have mentioned in the original post that I would do that, but I wondered whether perhaps there was some good way of covering both at the same time, or with the same explanation. $\endgroup$– user1205901 - Слава УкраїніMay 12, 2015 at 12:16
Add a comment
|