## Designing experiments and analyzing data: a model comparison perspectiveRather than presenting each experimental design in terms of a set of computational formulas to be used only for that design, this book uses a model comparison approach to present a few basic formulas that can be applied with the same underlying logic to every experimental design Once students understand the underlying principles of this general approach and master a few basic formulas introduced early on, they are able to view specific formulas for each individual design they encounter as a special case of more general formulas. In short, they are able to understand the logic that should guide their choice of a technique for a particular design. |

### What people are saying - Write a review

#### Review: Designing Experiments and Analyzing Data: A Model Comparison Perspective

User Review - Ohud - GoodreadsIt is a great book, but coming from a technical background I had a challenge to really have a deep understanding of it, classes help actually to discuss it with other classmates. Read full review

#### Review: Designing Experiments and Analyzing Data: A Model Comparison Perspective

User Review - Melissa Maxwell Davis - GoodreadsI'll admit that I'm a little biased, since my dad is the author. ;) Read full review

### Contents

Threats to the Validity of Inferences from Experiments | 25 |

Introduction to the Fisher Tradition | 38 |

Exercises | 56 |

Copyright | |

44 other sections not shown

### Other editions - View all

### Common terms and phrases

absent adjusted analysis ANCOVA angle appropriate assume average between-subjects design between-subjects factor biofeedback Bonferroni calculated cell means Chapter condition consider correlation covariate critical F critical value degrees of freedom denominator degrees dependent variable deviation diet difference discussed distribution drug equal Equation F statistic F test factorial design formula full model grand mean Kruskal-Wallis test linear trend main effect marginal means matrix mean square method mixed-model approach multivariate approach multivariate test noise null hypothesis number of subjects observed F value obtained omnibus test one-way design orthogonal pairwise comparisons parameters performed population means predicted present procedure quadratic trend random factor regression restricted model sample means sample sizes scores separate error term simple effects simply slope specific split-plot design squared errors statistically significant sum of squares test statistic therapy treatment effect Type I error Type III sum within-subjects design within-subjects factor yields zero