Biostatistical Design and Analysis Using R -- Contents -- Preface -- R quick reference card -- General key to statistical methods -- 1 Introduction to R -- 1.1 Why R? -- 1.2 Installing R -- 1.2.1 Windows -- 1.2.2 Unix/Linux -- 1.2.3 MacOSX -- 1.3 The R environment -- 1.3.1 The console (command line) -- 1.4 Object names -- 1.5 Expressions, Assignment and Arithmetic -- 1.6 R Sessions and workspaces -- 1.6.1 Cleaning up -- 1.6.2 Workspaces -- 1.6.3 Current working directory -- 1.6.4 Quitting R -- 1.7 Getting help -- 1.8 Functions -- 1.9 Precedence -- 1.10 Vectors - variables -- 1.10.1 Regular or patterned sequences -- 1.10.2 Character vectors -- 1.10.3 Factors -- 1.11 Matrices, lists and data frames -- 1.11.1 Matrices -- 1.11.2 Lists -- 1.11.3 Data frames - data sets -- 1.12 Object information and conversion -- 1.12.1 Object information -- 1.12.2 Object conversion -- 1.13 Indexing vectors, matrices and lists -- 1.13.1 Vector indexing -- 1.13.2 Matrix indexing -- 1.13.3 List indexing -- 1.14 Pattern matching and replacement (character search and replace) -- 1.14.1 grep - pattern searching -- 1.14.2 regexpr - position and length of match -- 1.14.3 gsub - pattern replacement -- 1.15 Data manipulation -- 1.15.1 Sorting -- 1.15.2 Formatting data -- 1.16 Functions that perform other functions repeatedly -- 1.16.1 Along matrix margins -- 1.16.2 By factorial groups -- 1.16.3 By objects -- 1.17 Programming in R -- 1.17.1 Grouped expressions -- 1.17.2 Conditional execution - if and ifelse -- 1.17.3 Repeated execution - looping -- 1.17.4 Writing functions -- 1.18 An introduction to the R graphical environment -- 1.18.1 The plot() function -- 1.18.2 Graphical devices -- 1.18.3 Multiple graphics devices -- 1.19 Packages -- 1.19.1 Manual package management -- 1.19.2 Loading packages -- 1.20 Working with scripts -- 1.21 Citing R in publications.

1.22 Further reading -- 2 Datasets -- 2.1 Constructing data frames -- 2.2 Reviewing a data frame - fix() -- 2.3 Importing (reading) data -- 2.3.1 Import from text file -- 2.3.2 Importing from the clipboard -- 2.3.3 Import from other software -- 2.4 Exporting (writing) data -- 2.5 Saving and loading of R objects -- 2.6 Data frame vectors -- 2.6.1 Factor levels -- 2.7 Manipulating data sets -- 2.7.1 Subsets of data frames - data frame indexing -- 2.7.2 The %in% matching operator -- 2.7.3 Pivot tables and aggregating datasets -- 2.7.4 Sorting datasets -- 2.7.5 Accessing and evaluating expressions within the context a dataframe -- 2.7.6 Reshaping dataframes -- 2.8 Dummy data sets - generating random data -- 3 Introductory statistical principles -- 3.1 Distributions -- 3.1.1 The normal distribution -- 3.1.2 Log-normal distribution -- 3.2 Scale transformations -- 3.3 Measures of location -- 3.4 Measures of dispersion and variability -- 3.5 Measures of the precision of estimates - standard errors and confidence intervals -- 3.6 Degrees of freedom -- 3.7 Methods of estimation -- 3.7.1 Least squares (LS) -- 3.7.2 Maximum likelihood (ML) -- 3.8 Outliers -- 3.9 Further reading -- 4 Sampling and experimental design with R -- 4.1 Random sampling -- 4.2 Experimental design -- 4.2.1 Fully randomized treatment allocation -- 4.2.2 Randomized complete block treatment allocation -- 5 Graphical data presentation -- 5.1 The plot() function -- 5.1.1 The type parameter -- 5.1.2 The xlim and ylim parameters -- 5.1.3 The xlab and ylab parameters -- 5.1.4 The axes and ann parameters -- 5.1.5 The log parameter -- 5.2 Graphical Parameters -- 5.2.1 Plot dimensional and layout parameters -- 5.2.2 Axis characteristics -- 5.2.3 Character sizes -- 5.2.4 Line characteristics -- 5.2.5 Plotting character parameter - pch -- 5.2.6 Fonts -- 5.2.7 Text orientation and justification.

5.2.8 Colors -- 5.3 Enhancing and customizing plots with low-level plotting functions -- 5.3.1 Adding points - points() -- 5.3.2 Adding text within a plot - text() -- 5.3.3 Adding text to plot margins - mtext() -- 5.3.4 Adding a legend - legend() -- 5.3.5 More advanced text formatting -- 5.3.6 Adding axes - axis() -- 5.3.7 Adding lines and shapes within a plot -- 5.4 Interactive graphics -- 5.4.1 Identifying points - identify() -- 5.4.2 Retrieving coordinates - locator() -- 5.5 Exporting graphics -- 5.5.1 Postscript - poscript() and pdf() -- 5.5.2 Bitmaps - jpeg() and png() -- 5.5.3 Copying devices - dev.copy() -- 5.6 Working with multiple graphical devices -- 5.7 High-level plotting functions for univariate (single variable) data -- 5.7.1 Histogram -- 5.7.2 Density functions -- 5.7.3 Q-Q plots -- 5.7.4 Boxplots -- 5.7.5 Rug charts -- 5.8 Presenting relationships -- 5.8.1 Scatterplots -- 5.9 Presenting grouped data -- 5.9.1 Boxplots -- 5.9.2 Boxplots for grouped means -- 5.9.3 Interaction plots - means plots -- 5.9.4 Bargraphs -- 5.9.5 Violin plots -- 5.10 Presenting categorical data -- 5.10.1 Mosaic plots -- 5.10.2 Association plots -- 5.11 Trellis graphics -- 5.11.1 scales() parameters -- 5.12 Further reading -- 6 Simple hypothesis testing - one and two population tests -- 6.1 Hypothesis testing -- 6.2 One- and two-tailed tests -- 6.3 t-tests -- 6.4 Assumptions -- 6.5 Statistical decision and power -- 6.6 Robust tests -- 6.7 Further reading -- 6.8 Key for simple hypothesis testing -- 6.9 Worked examples of real biological data sets -- 7 Introduction to Linear models -- 7.1 Linear models -- 7.2 Linear models in R -- 7.3 Estimating linear model parameters -- 7.3.1 Linear models with factorial variables -- 7.3.2 Linear model hypothesis testing -- 7.4 Comments about the importance of understanding the structure and parameterization of linear models.

8 Correlation and simple linear regression -- 8.1 Correlation -- 8.1.1 Product moment correlation coefficient -- 8.1.2 Null hypothesis -- 8.1.3 Assumptions -- 8.1.4 Robust correlation -- 8.1.5 Confidence ellipses -- 8.2 Simple linear regression -- 8.2.1 Linear model -- 8.2.2 Null hypotheses -- 8.2.3 Assumptions -- 8.2.4 Multiple responses for each level of the predictor -- 8.2.5 Model I and II regression -- 8.2.6 Regression diagnostics -- 8.2.7 Robust regression -- 8.2.8 Power and sample size determination -- 8.3 Smoothers and local regression -- 8.4 Correlation and regression in R -- 8.5 Further reading -- 8.6 Key for correlation and regression -- 8.7 Worked examples of real biological data sets -- 9 Multiple and curvilinear regression -- 9.1 Multiple linear regression -- 9.2 Linear models -- 9.3 Null hypotheses -- 9.4 Assumptions -- 9.5 Curvilinear models -- 9.5.1 Polynomial regression -- 9.5.2 Nonlinear regression -- 9.5.3 Diagnostics -- 9.6 Robust regression -- 9.7 Model selection -- 9.7.1 Model averaging -- 9.7.2 Hierarchical partitioning -- 9.8 Regression trees -- 9.9 Further reading -- 9.10 Key and analysis sequence for multiple and complex regression -- 9.11 Worked examples of real biological data sets -- 10 Single factor classification (ANOVA) -- 10.0.1 Fixed versus random factors -- 10.1 Null hypotheses -- 10.2 Linear model -- 10.3 Analysis of variance -- 10.4 Assumptions -- 10.5 Robust classification (ANOVA) -- 10.6 Tests of trends and means comparisons -- 10.7 Power and sample size determination -- 10.8 ANOVA in R -- 10.9 Further reading -- 10.10 Key for single factor classification (ANOVA) -- 10.11 Worked examples of real biological data sets -- 11 Nested ANOVA -- 11.1 Linear models -- 11.2 Null hypotheses -- 11.2.1 Factor A - the main treatment effect -- 11.2.2 Factor B - the nested factor -- 11.3 Analysis of variance.

11.4 Variance components -- 11.5 Assumptions -- 11.6 Pooling denominator terms -- 11.7 Unbalanced nested designs -- 11.8 Linear mixed effects models -- 11.9 Robust alternatives -- 11.10 Power and optimisation of resource allocation -- 11.11 Nested ANOVA in R -- 11.11.1 Error strata (aov) -- 11.11.2 Linear mixed effects models (lme and lmer) -- 11.12 Further reading -- 11.13 Key for nested ANOVA -- 11.14 Worked examples of real biological data sets -- 12 Factorial ANOVA -- 12.1 Linear models -- 12.2 Null hypotheses -- 12.2.1 Model 1 - fixed effects -- 12.2.2 Model 2 - random effects -- 12.2.3 Model 3 - mixed effects -- 12.3 Analysis of variance -- 12.3.1 Quasi F-ratios -- 12.3.2 Interactions and main effects tests -- 12.4 Assumptions -- 12.5 Planned and unplanned comparisons -- 12.6 Unbalanced designs -- 12.6.1 Missing observations -- 12.6.2 Missing combinations - missing cells -- 12.7 Robust factorial ANOVA -- 12.8 Power and sample sizes -- 12.9 Factorial ANOVA in R -- 12.10 Further reading -- 12.11 Key for factorial ANOVA -- 12.12 Worked examples of real biological data sets -- 13 Unreplicated factorial designs - randomized block and simple repeated measures -- 13.1 Linear models -- 13.2 Null hypotheses -- 13.2.1 Factor A - the main within block treatment effect -- 13.2.2 Factor B - the blocking factor -- 13.3 Analysis of variance -- 13.4 Assumptions -- 13.4.1 Sphericity -- 13.4.2 Block by treatment interactions -- 13.5 Specific comparisons -- 13.6 Unbalanced un-replicated factorial designs -- 13.7 Robust alternatives -- 13.8 Power and blocking efficiency -- 13.9 Unreplicated factorial ANOVA in R -- 13.10 Further reading -- 13.11 Key for randomized block and simple repeated measures ANOVA -- 13.12 Worked examples of real biological data sets -- 14 Partly nested designs: split plot and complex repeated measures -- 14.1 Null hypotheses.

14.1.1 Factor A - the main between block treatment effect.