Practical Data Analysis for Designed ExperimentsPlacing data in the context of the scientific discovery of knowledge through experimentation, Practical Data Analysis for Designed Experiments examines issues of comparing groups and sorting out factor effects and the consequences of imbalance and nesting, then works through more practical applications of the theory. Written in a modern and accessible manner, this book is a useful blend of theory and methods. Exercises included in the text are based on real experiments and real data. |
Contents
Collaboration in Science | 21 |
Experimental Design | 35 |
Working with Groups of Data | 47 |
Comparing Several Means | 71 |
Multiple Comparisons of Means | 89 |
Sorting out Effects with Data | 105 |
Balanced Experiments | 125 |
Model Selection | 145 |
Parallel Lines | 255 |
Multiple Responses | 275 |
Deciding on Fixed or Random Effects | 295 |
General Random Models | 313 |
Mixed Effects Models | 327 |
Nesting Experimental Units | 335 |
Split Plot Design | 357 |
General Nested Designs | 369 |
Dealing with Imbalance | 159 |
Missing Cells | 177 |
Linear Models Inference | 195 |
Questioning Assumptions | 207 |
Comparisons with Unequal Variance | 221 |
Getting Free from Assumptions | 229 |
Regressing with Factors | 239 |
Other editions - View all
Common terms and phrases
additive model adjusted analysis of variance anova approach appropriate assigned assumptions balanced block cell means chapter Cloning comparing completely randomized design confidence intervals consider contrasts correlation covariate critical value data analysis degrees of freedom distribution E(MS effects model estimable functions examine Example expected mean squares experiment experimental units F test factor combinations factor levels Figure fixed effects grand mean group means inference interaction plots interpretation Latin square linear constraints main effects marginal means matrix mean response methods multiple comparisons multivariate non-centrality parameter null hypothesis one-factor orthogonal overall test p-value partition pivot statistic ploidy population problem proc quadratic forms random effects random model regression REML repeated measures replication S-Plus sample sizes SCHEFFÉ scientist Section set-to-zero significant source df SS split plot design ẞj statistician sub-sampling sum of squares treatment structure two-factor Type unbalanced variance components variation whole plot yield Yijk