Sample Size and ANOVA
The sample size situation with regard to ANOVA is typically
analogous to issue as explored in the context of a t -test:
Sample Size and ANOVA
The sample size situation with regard to ANOVA is typically
analogous to issue as explored
Factorial Design
To this point we have only considered experiments with a single
factor at g levels. The true power and utility of the ANOVA
concept becomes clear, though, in the context of experiments in
which there are two or more factors, each at sever
Sums-of-Squares in Factorial Structure
Recall that for a single-factor ANOVA with g treatment levels
SST = SSTrt + SSE
Sums-of-Squares in Factorial Structure
Recall that for a single-factor ANOVA with g treatment levels
SST = SSTrt + SSE
This occurs becau
A Few Odds and Ends
Perhaps surprisingly, the model we choose to explain the response
in terms of the factors might depend on the scale on which we
measure the response.
A Few Odds and Ends
Perhaps surprisingly, the model we choose to explain the response
Examining Interaction
Suppose now that we have examined the ANOVA table for a twoor many-way factorial design and have determined that there is an
interaction eect. The next step is to examine the data to
determine how the interaction works.
Examining Int
Unbalanced Design
Unbalanced design in an ANOVA means that the number of
observations (experimental units) at each factor-level combination
are unequal. In a one-way ANOVA this is not a particular problem,
although it makes the construction of orthogonal
Random Eects
To this point we have considered models in which treatment
eects are parameters, i.e. xed unvarying numbers. Our
principle focus has been to decide whether those numbers are
signicantly dierent than zero and, if so, perhaps to nd interval
est