MANE 6313
Week 8, Module D
Student Learning Outcome
- Select an appropriate experimental design with one or more factors,
- Select an appropriate model with one or more factors,
- Evaluate statistical analyses of experimental designs,
- Assess the model adequacy of any experimental design, and
- Interpret model results.
Module Learning Outcome
Analyze a two-level factorial design in two blocks using R.
Problem 7.5 (9th Edition)
Data Frame Creation
Data Frame Creation, continued
Half Normal Plot
- Notice that \(\alpha\) was set to 0.25
Model 2
Model 3
Model 4
Modeling Summary
-
Since single replicate was created, half normal plot is the first step
-
Set \(\alpha\) to 0.25 to match break in plot
-
Try to balance reasonable values of \(\alpha\) with shape of half normal plot
-
Prefer to use larger \(\alpha\) when starting to analyze unreplicated experiments
-
-
For Model 2, there as choice of using \(\alpha\)=0.05 or \(\alpha=0.1\). My preference is to make small changes in model so selected \(\alpha=0.1\). In retrospective, could have gone from model 2 to model using \(\alpha\)=0.05
R Demonstration
- Creating data frame with two blocks