The New Statistics with R: An Introduction for Biologists

Cover
The New Statistics with R: An Introduction for Biologists
Copyright
Dedication
Acknowledgements
Contents
1: Introduction
	1.1 Introduction to the second edition
	1.2 The aim of this book
	1.3 Changes in the second edition
	1.4 The R programming language for statistics and graphics
	1.5 Scope
	1.6 What is not covered
	1.7 The approach
	1.8 The new statistics?
	1.9 Getting started
	1.10 References
2: Motivation
	2.1 A matter of life and death
	2.2 Summary: Statistics
	2.3 Summary: R
	2.4 References
3: Description
	3.1 Introduction
		3.1.1 r packages
	3.2 Darwin’s maize pollination data
		3.2.1 know your data
		3.2.2 summarizing and describing data
		3.2.3 comparing groups
	3.3 Summary: Statistics
	3.4 Summary: R
	3.5 References
4: Reproducible Research
	4.1 The reproducibility crisis
		4.1.1 R packages
	4.2 R scripts
	4.3 Analysis notebooks
	4.4 R Markdown
	4.5 Summary: Statistics
	4.6 Summary: R
	4.7 References
5: Estimation
	5.1 Introduction
		5.1.1 R Packages
	5.2 Quick tests
	5.3 Differences between groups
	5.4 Standard deviations and standard errors
	5.5 The normal distribution and the central limit theorem
	5.6 Confidence intervals
	5.7 Summary: Statistics
	5.8 Summary: R
	Appendix 5a: R code for Fig. 5.1
6: Linear Models
	6.1 Introduction
		6.1.1 R Packages
	6.2 A linear-model analysis for comparing groups
	6.3 Standard error of the difference
	6.4 Confidence intervals
	6.5 Answering Darwin’s question
	6.6 Relevelling to get the other treatment mean and standard error
	6.7 Assumption checking
		6.7.1 Normality
		6.7.2 Equal Variance
	6.8 Summary: Statistics
	6.9 Summary: R
	6.10 Reference
	Appendix 6a: R graphics
	Appendix 6b: Robust linear models
	Appendix 6c: Exercise
7: Regression
	7.1 Introduction
		7.1.1 R Packages
	7.2 Linear regression
	7.3 The Janka timber hardness data
	7.4 Correlation
	7.5 Linear regression in R
	7.6 Assumptions
	7.7 Summary: Statistics
	7.8 Summary: R
	7.9 Reference
	Appendix 7a: R graphics
	Appendix 7b: Least squares linear regression
8: Prediction
	8.1 Introduction
		8.1.1 R Packages
	8.2 Predicting timber hardness from wood density
	8.3 Confidence intervals and prediction intervals
	8.4 Summary: Statistics
	8.5 Summary: R
9: Testing
	9.1 Significance testing: Time for t
		9.1.1 R Packages
	9.2 Student’s t-test: Darwin’s maize
	9.3 Summary: Statistics
	9.4 Summary: R
	9.5 References
10: Intervals
	10.1 Comparisons using estimates and intervals
		10.1.1 R Packages
	10.2 Estimation-based analysis
	10.3 Descriptive statistics
		10.3.1 Error Bars
		10.3.2 Standard-deviation Error Bars
	10.4 Inferential statistics
		10.4.1 Standard Error Bars
		10.4.2 Confidence Intervals
		10.4.3 Confidence Intervals For Differences Between Means
		10.4.4 Least Significant Differences
		10.4.5 Multi-interval Plots
		10.4.6 Prediction Intervals
	10.5 Relating different types of interval and error bar
		10.5.1 Interpreting Confidence Intervals
		10.5.2 Point Estimates And Confidence Intervals For Research Synthesis And Meta-analysis
	10.6 Summary: Statistics
	10.7 Summary: R
	10.8 References
11: Analysis of Variance
	11.1 ANOVA tables
		11.1.1 R packages
	11.2 ANOVA tables: Darwin’s maize
	11.3 Hypothesis testing: F-values
	11.4 Two-way ANOVA
	11.5 Summary
	11.6 Reference
12: Factorial Designs
	12.1 Introduction
		12.1.1 R Packages
	12.2 Factorial designs
	12.3 Comparing three or more groups
	12.4 Two-way ANOVA (no interaction)
	12.5 Additive treatment effects
	12.6 Interactions: Factorial ANOVA
		12.6.1 Factorial Anova In R
	12.7 Summary: Statistics
	12.8 Summary: R
	12.9 References
	Appendix 12a: Code for Fig. 12.3
13: Analysis of Covariance
	13.1 ANCOVA
		13.1.1 R Packages
	13.2 The agricultural pollution data
		13.2.1 Warning: R Regenerates!
	13.3 ANCOVA with water stress and low-level ozone
	13.4 Interactions in ANCOVA
	13.5 General linear models
	13.6 Summary
	13.7 References
14: Linear Model Complexities
	14.1 Introduction
	14.2 Analysis of variance for balanced designs
	14.3 Analysis of variance with unbalanced designs
	14.4 ANOVA tables versus coefficients: When F and t can disagree
	14.5 Marginality of main effects and interactions
	14.6 Summary
	14.7 References
15: Generalized Linear Models
	15.1 GLMs
	15.2 The trouble with transformations
	15.3 The Box–Cox power transform
	15.4 Generalized Linear Models in R
	15.5 Summary: Statistics
	15.6 Summary: R
	15.7 References
16: GLMs for Count Data
	16.1 Introduction
	16.2 GLMs for count data
	16.3 Quasi-maximum likelihood
	16.4 Summary
17: Binomial GLMs
	17.1 Binomial counts and proportion data
	17.2 The beetle data
	17.3 GLM for binomial counts
	17.4 Alternative link functions
	17.5 Summary: Statistics
	17.6 Summary: R
	17.7 Reference
18: GLMs for Binary Data
	18.1 Binary data
		18.1.1 R Packages
	18.2 The wells data set for the binary GLM example
	18.3 Centering
	18.4 Summary
	18.5 References
19: Conclusions
	19.1 Introduction
	19.2 A binomial GLM analysis of the Challenger binary data
	19.3 Recommendations
	19.4 Where next?
	19.5 Further reading
	19.6 The R café
	19.7 References
20: A Very Short Introduction to R
	20.1 Installing R
	20.2 Installing RStudio
	20.3 R packages
	20.4 The R language
		20.4.1 Functions
		20.4.2 Arguments
		20.4.3 Objects
		20.4.4 Dataframes
		20.4.5 Graphics
Index