Better experimental design and statistical analysis make for more robust science. A thorough understanding of modern statistical methods can mean the difference between discovering and missing crucial results and conclusions in your research, and can shape the course of your entire research career. With Applied Statistics, Barry Glaz and Kathleen M. Yeater have worked with a team of expert authors to create a comprehensive text for graduate students and practicing scientists in the agricultural, biological, and environmental sciences. The contributors cover fundamental concepts and methodologies of experimental design and analysis, and also delve into advanced statistical topics, all explored by analyzing real agronomic data with practical and creative approaches using available software tools. IN PRESS! This book is being published according to the “Just Published” model, with more chapters to be published online as they are completed.
Author(s): Barry Glaz, Kathleen M. Yeater
Edition: 1
Publisher: ACSESS
Year: 2020
Language: English
Commentary: Appendices A, B and C included
Pages: 672
Capa_digital_art_x4_colored_toned_light_ai.pdf (p.1)
Ch0.pdf (p.2-21)
Ch1.pdf (p.22-39)
CHAPTER 1: ERRORS IN STATISTICAL DECISION MAKING
Statistical Decisions in Agronomic and Environmental Research Are Often Framed as Null Hypothesis Tests
Effect Size
The Noncentral Distribution
Errors in Statistical Decision Making
Example
Rethinking Standard Operating Procedure:
Relative Importance of α and β Errors
Reasons for Lack of Significance Not Reported
Complicated Experimental Designs with Multiple Possible Hypotheses Are Frequently Under-Powered
Errors in Split Plot Experiments
Recommendations—So What Is a Person to Do?
Step-by-Step Recommendations
Summary
Key Learning Points
Review Questions
True or False:
Exercises
References
Ch2.pdf (p.40-73)
CHAPTER 2: ANALYSIS OF VARIANCE AND HYPOTHESIS TESTING
The ANOVA Process
History of Analysis of Variance
Planning an Experiment
The Linear Model
Fixed and Random Effects
Fixed, Mixed, and Random Models
ANOVA COMPONENTS
Sources of Variation
Degrees of Freedom
Sums of Squares
Mean Squares
F-values
P-value
Contrasts and Multiple Comparison Procedures
Case Study: The Story of Statbean: From Discovery to Field Testing
Introduction
Research Objectives
Experimental Description and Design– Randomization and Replication of Treatments
Sampling Description and Design–Measuring Dependent Variables
Preparing, Correcting, and Knowing the Data
Descriptive Summary of the Data
ANOVA by Location
ANOVA by Location SAS code for PROC MIXED
Residual Plots
Results
Planned and Multiple Pairwise Comparisons
ANOVA Combined Over Fixed Locations
SAS Code– PROC MIXED- pH Combined Over Locations
Checking for Heterogeneity of Variance
ANOVA Results– Combined over Locations
Planned and Multiple Pairwise Comparison
Presentation of ANOVA Results
Conclusions
Key Learning Points
Review Questions (T/F)
Exercises
References
Ch3.pdf (p.74-93)
CHAPTER 3: BLOCKING PRINCIPLES FOR BIOLOGICAL EXPERIMENTS
Randomization
Blocking
Concepts of Blocking
Complete Block Designs
Incomplete Block Designs
Split-Plot Blocking Patterns
Change-Over (Crossover) Designs
The Special Needs of Glasshouse and Growth Chamber Experimentation
Fixed vs. Random Effects
To Pool or Not to Pool
The Value of Retrospective Analyses
Conclusions
Key Learning Points
Exercises
References
Ch4.pdf (p.94-104)
CHAPTER 4: POWER AND REPLICATION—DESIGNING POWERFUL EXPERIMENTS
Concepts of Replication
What is an Experimental Unit?
Replication on Multiple Scales
Replication and Power
Conclusions
Key Learning Points
Exercises
References
Ch5.pdf (p.105-125)
CHAPTER 5: MULTIPLE COMPARISON PROCEDURES: THE INS AND OUTS
Types of Statistical Error
Traditional Hypothesis Testing Scenario
How To Get The Statistical Analysis Wrong
Garlic Example
Alfalfa Example
Ideas, Hypotheses, and Contrasts of Various Types
What If There Are No Prior Ideas?
Multiple Comparison Procedures
Ordering of Multiple Comparison Procedures by Level of Conservatism
What Is the Natural Unit?
Inconsistency of Multiple Comparison Procedures
Goldilocks and the Four Bears
The Inconsistency of Other Multiple Comparison Procedures
The Significance Level and Power of Some Well-Known Multiple Comparison Procedures
Power Analysis
The Practical Solution
Advantages of Using the Unrestricted LSD Procedure
General Contrasts and Report Writing
Example of Usage of the Practical Solution in Research
Conclusions
Key Learning Points
Exercises
Acknowledgments
References
Ch6.pdf (p.126-195)
CHAPTER 6: LINEAR REGRESSION TECHNIQUES
Historical Background
Some Aspects of Planning Experiments for Linear Regression
The Basic Idea
Linear vs. Nonlinear Regression
The Simple Linear Regression
The Model
Statistical Inferences
Results of Examples 1 to 4
Diagnostics of the Residuals
The Problem of Regressand Transformation—Reconsideration of Example 4
Multiple Linear Regression
Specifics of the Multiple Linear Regression Model Compared to the Simple Linear Regression Model
Sequential and Partial Evaluation of Regressors
Techniques for Model Selection
The No-Intercept Problem in Simple and Multiple Linear Regression
Extensions of the Linear Regression Models
The Linear Mixed Model
Example 6 Continued.
Example 7 (SAS Code in Appendix)
Example 3 (Modified) (SAS Code in Appendix)
Example 8 (SAS Code in Appendix)
Concluding Remarks
Key Learning Points
Review Questions
Exercises
QUESTIONS
ACKNOWLEDGMENTS
References
Ch7.pdf (p.196-218)
CHAPTER 7: ANALYSIS AND INTERPRETATION OF INTERACTIONS OF FIXED AND RANDOM EFFECTS
Looking for the Best Model
Experimental Data
The Complete Model: Analyzing Main Effects and Their Two-Way, Three-Way, and Four-Way Interactions
Interpreting the Year × Soil × N interaction
Interpreting Year × Soil × P Interaction
Interpreting Soil × N × P interaction
Interpreting the Adjusted Means
Mixed Models
Practical Recommendations Related to the Fixed vs. Random
Effects Debate
Example 1
Example 2
Conclusions
Key Learning Points
Review Questions
Acknowledgments
References
Ch8.pdf (p.219-252)
CHAPTER 8: THE ANALYSIS OF COMBINED EXPERIMENTS
A Linear Model for Qualitative Treatments in a Combined Experiment
The Choice of Fixed or Random Effect in General
Example: Oat Cultivar Trial
Choosing Whether to Subdivide the Treatment by Environment Interaction by Environment
Choosing Whether to Subdivide the Treatment by Environment Interaction by Treatment
Choosing Whether or Not to Assume Equal Residual Variances for All Environments
Summary
Key Learning Points:
Review Questions (T/F)
Exercises
References
Ch9.pdf (p.253-295)
CHAPTER 9: ANALYSIS OF COVARIANCE
Abstract
Introduction
Description of the ANCOVA Model
Summary
Key Learning Points
Review Questions
Questions To Be Considered For Each Data Analysis Exercise
Acknowledgments
References
Ch10.pdf (p.296-314)
CHAPTER 10: ANALYSIS OF REPEATED MEASURES FOR THE BIOLOGICAL AND AGRICULTURAL SCIENCES
Abstract
Linear Mixed Models
Variance–Covariance Structures
Conclusions
Key Learning Points
Review Questions
Exercises
References
Ch11.pdf (p.315-333)
CHAPTER 11: THE DESIGN AND ANALYSIS OF LONG-TERM ROTATION EXPERIMENTS
Analysis
Summary
Acknowledgments
Key Learning Points
Review Questions
Exercises
References
Ch12.pdf (p.334-359)
CHAPTER 12: SPATIAL ANALYSIS OF FIELD EXPERIMENTS
Practical Considerations
Experimental Design
Spatial Model
Example 1
Summary
Key Learning Points
Review Questions
Exercises
References
Ch13.pdf (p.360-384)
CHAPTER 13: AUGMENTED DESIGNS-EXPERIMENTAL DESIGNS IN WHICH ALL TREATMENTS ARE NOT REPLICATED
INTRODUCTION
Precision
Spatial Adjustment
Analysis
Example 2: Augmented Design with Systematic Checks in a Latin-Square Arrangement
Summary
Key Learning Points
Review Questions
Exercises
References
Ch14.pdf (p.385-413)
CHAPTER 14: MULTIVARIATE METHODS FOR AGRICULTURAL RESEARCH
Questions and Methods: A Field Guide
Begin Data Analysis by Understanding the Nature of the Data
Example Data, Exploration, and Multivariate Applications
Data Exploration
Summary and Final Remarks
Key Learning Points
Review Questions
Exercises
References
Ch15.pdf (p.414-460)
CHAPTER 15: NONLINEAR REGRESSION MODELS AND APPLICATIONS
Nonlinear Regression Model
Why Should We Use Nonlinear Models?
Example 1: Maize Biomass Accumulation
Comparing Alternative Models
Example 2: Extended Application
Summary
Key Learning Points
Review Questions
Exercises
References
Ch16.pdf (p.461-521)
CHAPTER 16: ANALYSIS OF NON-GAUSSIAN DATA
Abstract
A Brief History of Non-Gaussian Data Analysis in the Plant Sciences
Introduction to GLMM – Key Concepts and Requirements for Implementation
Some Things You Need to Know About Probability Distributions
What Would Fisher Do?
Gentle Introduction to the GLMM
Key Issues – Data Versus Model Scale; Conditional Versus Marginal Model
Model Versus Data Scale
What to Include in the Linear Predictor
Conditional and Marginal
GLIMMIX Syntax Basics
The Examples
Example 1 – A Randomized Complete Block Design with Binomial Data
Relevant Output
Example 2 – Split-Plot experiment with Count Data
Example 3 – A Multi-Location (Blocked) Design with Categorical (Multinomial) Data
Relevant Output I: Solution
Example 4 – A Repeated Measures Experiment with Count Data
Description of the Study and Objectives
Key Learning Points
Review Questions
REFERENCES
AppendixA.pdf (p.522-650)
AppendixB_onlineonly.pdf (p.651-1207)
AppendixC_onlineonly.pdf (p.1208-1440)
