Experimental Design and Data Analysis for Biologists

This essential textbook is designed for students or researchers in biology who need to design experiments, sampling programs, or analyze resulting data. The text begins with a revision of estimation and hypothesis testing methods, before advancing to the analysis of linear and generalized linear models. The chapters include such topics as linear and logistic regression, simple and complex ANOVA models, log-linear models, and multivariate techniques. The main analyses are illustrated with many examples from published papers and an extensive reference list to both the statistical and biological literature is also included. The book is supported by a web-site that provides all data sets, questions for each chapter and links to software. Cover Page 1 Book Info 3 Title Page 5 ISBN 0521811287 6 Contents (with page links) 7 1 Introduction 7 2 Estimation 7 3 Hypothesis testing 8 4 Graphical exploration of data 8 5 Correlation and regression 9 6 Multiple and complex regression 9 7 Design and power analysis 10 8 Comparing groups or treatments – analysis of variance 10 9 Multifactor analysis of variance 11 10 Randomized blocks and simple repeated measures: unreplicated two factor designs 12 11 Split-plot and repeated measures designs: partly nested analyses of variance 12 12 Analyses of covariance 13 13 Generalized linear models and logistic regression 14 14 Analyzing frequencies 14 15 Introduction to multivariate analyses 14 16 Multivariate analysis of variance and discriminant analysis 15 17 Principal components and correspondence analysis 15 18 Multidimensional scaling and cluster analysis 16 19 Presentation of results 16 References, Index 16 Preface 17 In this book 17 Learning by example 18 This book is a bridge 18 Some acknowledgments 18 Chapter 1 Introduction 21 1.1 Scientific method 21 1.2 Experiments and other tests 25 1.3 Data, observations and variables 27 1.4 Probability 27 1.5 Probability distributions 29 Chapter 2 Estimation 34 2.1 Samples and populations 34 2.2 Common parameters and statistics 35 2.3 Standard errors and confidence intervals for the mean 37 2.4 Methods for estimating parameters 43 2.5 Resampling methods for estimation 45 2.6 Bayesian inference – estimation 47 Chapter 3 Hypothesis testing 52 3.1 Statistical hypothesis testing 52 3.2 Decision errors 62 3.3 Other testing methods 65 3.4 Multiple testing 68 3.5 Combining results from statistical tests 70 3.6 Critique of statistical hypothesis testing 71 3.7 Bayesian hypothesis testing 74 Chapter 4 Graphical exploration of data 78 4.1 Exploratory data analysis 78 4.2 Analysis with graphs 82 4.3 Transforming data 84 4.4 Standardizations 87 4.5 Outliers 88 4.6 Censored and missing data 88 4.7 General issues and hints for analysis 91 Chapter 5 Correlation and regression 92 5.1 Correlation analysis 92 5.2 Linear models 97 5.3 Linear regression analysis 98 5.4 Relationship between regression and correlation 126 5.5 Smoothing 127 5.6 Power of tests in correlation and regression 129 5.7 General issues and hints for analysis 130 Chapter 6 Multiple and complex regression 131 6.1 Multiple linear regression analysis 131 6.2 Regression trees 163 6.3 Path analysis and structural equation modeling 165 6.4 Nonlinear models 170 6.5 Smoothing and response surfaces 172 6.6 General issues and hints for analysis 173 Chapter 7 Design and power analysis 175 7.1 Sampling 175 7.2 Experimental design 177 7.3 Power analysis 184 7.4 General issues and hints for analysis 191 Chapter 8 Comparing groups or treatments – analysis of variance 193 8.1 Single factor (one way) designs 193 8.2 Factor effects 208 8.3 Assumptions 211 8.4 ANOVA diagnostics 214 8.5 Robust ANOVA 215 8.6 Specific comparisons of means 216 8.7 Tests for trends 222 8.8 Testing equality of group variances 223 8.9 Power of single factor ANOVA 224 8.10 General issues and hints for analysis 226 Chapter 9 Multifactor analysis of variance 228 9.1 Nested (hierarchical) designs 228 9.2 Factorial designs 241 9.3 Pooling in multifactor designs 280 9.4 Relationship between factorial and nested designs 281 9.5 General issues and hints for analysis 281 Chapter 10 Randomized blocks and simple repeated measures: unreplicated two factor designs 282 10.1 Unreplicated two factor experimental designs 282 10.2 Analyzing RCB and RM designs 288 10.3 Interactions in RCB and RM models 294 10.4 Assumptions 300 10.5 Robust RCB and RM analyses 304 10.6 Specific comparisons 305 10.7 Efficiency of blocking (to block or not to block?) 305 10.8 Time as a blocking factor 307 10.9 Analysis of unbalanced RCB designs 307 10.10 Power of RCB or simple RM designs 309 10.11 More complex block designs 310 10.12 Generalized randomized block designs 318 10.13 RCB and RM designs and statistical software 318 10.14 General issues and hints for analysis 319 Chapter 11 Split-plot and repeated measures designs: partly nested analyses of variance 321 11.1 Partly nested designs 321 11.2 Analyzing partly nested designs 329 11.3 Assumptions 338 11.4 Robust partly nested analyses 340 11.5 Specific comparisons 340 11.6 Analysis of unbalanced partly nested designs 342 11.7 Power for partly nested designs 343 11.8 More complex designs 343 11.9 Partly nested designs and statistical software 355 11.10 General issues and hints for analysis 357 Chapter 12 Analyses of covariance 359 12.1 Single factor analysis of covariance (ANCOVA) 359 12.2 Assumptions of ANCOVA 368 12.3 Homogeneous slopes 369 12.4 Robust ANCOVA 372 12.5 Unequal sample sizes (unbalanced designs) 373 12.6 Specific comparisons of adjusted means 373 12.7 More complex designs 373 12.8 General issues and hints for analysis 377 Chapter 13 Generalized linear models and logistic regression 379 13.1 Generalized linear models 379 13.2 Logistic regression 380 13.3 Poisson regression 391 13.4 Generalized additive models 392 13.5 Models for correlated data 395 13.6 General issues and hints for analysis 398 Chapter 14 Analyzing frequencies 400 14.1 Single variable goodness-of-fit tests 401 14.2 Contingency tables 401 14.3 Log-linear models 413 14.4 General issues and hints for analysis 420 Chapter 15 Introduction to multivariate analyses 421 15.1 Multivariate data 421 15.2 Distributions and associations 422 15.3 Linear combinations, eigenvectors and eigenvalues 425 15.4 Multivariate distance and dissimilarity measures 429 15.5 Comparing distance and/or dissimilarity matrices 434 15.6 Data standardization 435 15.7 Standardization, association and dissimilarity 437 15.8 Multivariate graphics 437 15.9 Screening multivariate data sets 438 15.10 General issues and hints for analysis 443 Chapter 16 Multivariate analysis of variance and discriminant analysis 445 16.1 Multivariate analysis of variance (MANOVA) 445 16.2 Discriminant function analysis 455 16.3 MANOVA vs discriminant function analysis 461 16.4 General issues and hints for analysis 461 Chapter 17 Principal components and correspondence analysis 463 17.1 Principal components analysis 463 17.2 Factor analysis 478 17.3 Correspondence analysis 479 17.4 Canonical correlation analysis 483 17.5 Redundancy analysis 486 17.6 Canonical correspondence analysis 487 17.7 Constrained and partial “ordination” 488 17.8 General issues and hints for analysis 491 Chapter 18 Multidimensional scaling and cluster analysis 493 18.1 Multidimensional scaling 493 18.2 Classification 508 18.3 Scaling (ordination) and clustering for biological data 511 18.4 General issues and hints for analysis 513 Chapter 19 Presentation of results 514 19.1 Presentation of analyses 514 19.2 Layout of tables 517 19.3 Displaying summaries of the data1 518 19.4 Error bars 524 19.5 Oral presentations 527 19.6 General issues and hints 530 References 531 Index 547 Publisher Description (unedited Publisher Data) An Essential Textbook For Any Student Or Researcher In Biology Needing To Design Experiments, Sampling Programs Or Analyse The Resulting Data. The Text Begins With A Revision Of Estimation And Hypothesis Testing Methods, Covering Both Classical And Bayesian Philosophies, Before Advancing To The Analysis Of Linear And Generalized Linear Models. Topics Covered Include Linear And Logistic Regression, Simple And Complex Anova Models (for Factorial, Nested, Block, Split-plot And Repeated Measures And Covariance Designs), And Log-linear Models. Multivariate Techniques, Including Classification And Ordination, Are Then Introduced. Special Emphasis Is Placed On Checking Assumptions, Exploratory Data Analysis And Presentation Of Results. The Main Analyses Are Illustrated With Many Examples From Published Papers And There Is An Extensive Reference List To Both The Statistical And Biological Literature. The Book Is Supported By A Web-site That Provides All Data Sets, Questions For Each Chapter And Links To Software. Introduction -- Estimation -- Hypothesis Testing -- Graphical Exploration Of Data -- Correlation And Regression -- Multiple And Complex Regression -- Design And Power Analysis -- Comparing Groups Or Treatments : Analysis Of Variance -- Multifactor Analysis Of Variance -- Randomized Blocks And Simple Repeated Measures : Unreplicated Two Factor Designs -- Split-plot And Repeated Measures Designs : Partly Nested Analyses Of Variance -- Analyses Of Covariance -- Generalized Linear Models And Logistic Regression -- Analyzing Frequencies -- Introduction To Multivariate Analyses -- Multivariate Analysis Of Variance And Discriminant Analysis -- Principal Components And Correspondence Analysis -- Multidimensional Scaling And Cluster Analysis -- Presentation Of Results. Gerry P. Quinn, Michael J. Keough. Includes Bibliographical References (p. [511]-526) And Index. An essential textbook for any biologist needing to design experiments, sampling programs or analyse the resulting data. Worked examples are used to illustrate the analyses and an extensive reference list provides links to the relevant biological and statistical literature. A supporting web-site contains datasets, questions and software links. Biologists and environmental scientists today must contend with the demands of keeping up with their primary field of specialization, and at the same time ensuring that their set of professional tools is current.