The New Statistics with R: An Introduction for BiologistsStatistical methods are a key tool for all scientists working with data, but learning the basic mathematical skills can be one of the most challenging components of a biologist's training. This accessible book provides a contemporary introduction to the classical techniques and modern extensions of linear model analysis: one of the most useful approaches in the analysis of scientific data in the life and environmental sciences. It emphasizes an estimation-based approach that accounts for recent criticisms of the over-use of probability values, and introduces alternative approaches using information criteria. Statistics are introduced through worked analyses performed in R, the free open source programming language for statistics and graphics, which is rapidly becoming the standard software in many areas of science and technology. These analyses use real data sets from ecology, evolutionary biology and environmental science, and the data sets and R scripts are available as support material. The book's structure and user friendly style stem from the author's 20 years of experience teaching statistics to life and environmental scientists at both the undergraduate and graduate levels. The New Statistics with R is suitable for senior undergraduate and graduate students, professional researchers, and practitioners in the fields of ecology, evolution, environmental studies, and computational biology. Supporting material for the book is available at the author's website: www.plantecol.org/contemporary-analysis-for-ecology/ |
Contents
1 | |
Analysis of Variance | 7 |
Students ttest | 35 |
4 Linear Regression | 51 |
5 Comparisons Using Estimates and Intervals | 67 |
6 Interactions | 83 |
ANCOVA | 101 |
8 Maximum Likelihood and Generalized Linear Models | 113 |
Other editions - View all
Common terms and phrases
2015 by Oxford Andy Hector 2015 ANOVA table approach approximately arm package average binomial count biodepth block box plots calculate Chapter coef.est coef.se Intercept coefficients complex count data Cross darwin Darwin’s maize data data frame data set default density Diversity2 estimates and intervals example explanatory variable F value Pr(>F F-L+ F+ L+ F+L+ fixed effects function output ggplot2 GLMMs graph grassland http://www.plantecol.org information criteria interaction janka likelihood linear model linear model analysis linear regression link function lme4 log-likelihood logit main effects maximum likelihood model selection multiple comparisons normal distribution normal least squares null hypothesis number of species overdispersion P-values pair plot Poisson Poisson distribution predicted produced Published 2015 qplot quantified quasi-maximum likelihood random effects ratio sample sizes significance tests slope standard error statistically significant summary t-test tion understorey variance components variation Well-watered Well-watered Well-watered zero