S-Plus code is presented for the graphical insight into, and the statistical analysis of, a two-treatment, two-period, two-treatment-sequence, or 2,2,2 crossover design. In this introductory section, we describe the 2,2,2 crossover design and its uses in the pharmaceutical industry with emphasis on food interaction studies. We also introduce a specific example and a dataset which will be used pedagogically throughout the chapter. In Section 7.2, we provide a brief introduction to data management in S-Plus demonstrating just enough manipulations to facilitate the graphical methods and data analyses which follow. Section 7.3 presents a series of graphs for the initial exploration and discovery stage of the analysis of the 2,2,2 crossover design. In Section 7.4, we perform the usual normal theory ANOVA and provide a clear and decision-oriented summary and inference plot. Section 7.5 presents several graphical tools for the “visualization of the ANOVA” and a subsequent model fit assessment, and we end with a summary in Section 7.6.
[1]
Beat Kleiner,et al.
Graphical Methods for Data Analysis
,
1983
.
[2]
Peter L. Brooks,et al.
Visualizing data
,
1997
.
[3]
Stephen Senn,et al.
Cross-over trials in clinical research
,
1993
.
[4]
16. Cross‐Over Trials in Clinical Research
,
1993
.
[5]
M. Braga,et al.
Exploratory Data Analysis
,
2018,
Encyclopedia of Social Network Analysis and Mining. 2nd Ed..
[6]
Andrew F. Siegel,et al.
Modern data analysis
,
1983
.
[7]
Bret Larget,et al.
Analysis of Categorical Data
,
2002
.
[8]
Gary G. Koch,et al.
Analysis of categorical data
,
1985
.
[9]
J. Matthews,et al.
Serial correlation in the design and analysis of crossover trials
,
1991
.
[10]
W. Cleveland.
Robust Locally Weighted Regression and Smoothing Scatterplots
,
1979
.
[11]
John M. Chambers,et al.
Graphical Methods for Data Analysis
,
1983
.