Ulteriori informazioni
Field research is a key element of scientific progress in agriculture and applied biology, bridging laboratory experiments in controlled conditions with on-farm trials conducted for demonstrative purposes. Like laboratory research, field studies must follow rigorous scientific protocols to yield reliable answers, while taking place in real-life settings to ensure practical relevance. The use of near-commercial-scale equipment and the need to address random variability, such as fertility gradients and pest invasions, have led to specific methodologies for experimental design and data analysis. These are seldom presented together in a concise, introductory format accessible to students and practitioners, and supported by real-world examples and reproducible code.
This book addresses that gap by focusing on small-plot field experiments evaluating genotypes, agronomic practices, pesticides, and plant protection methods. Using a 'learn-by-doing' approach, it presents selected examples and case studies, along with hands-on exercises. R is used as the primary tool for data analysis, with reproducible code snippets included. While not exhaustive, the book is tailored for a 6 ECTS introductory biometry course (approx. 54 hours), aimed at master s and PhD students in agriculture and biology, and practitioners seeking foundational knowledge. No prior statistics background is required, though basic familiarity with R is assumed; an introductory appendix is provided for beginners.
Topics include: (i) experimental design and layout types; (ii) descriptive statistics; (iii) model fitting (ANOVA, regression, mixed models); (iv) stochastic models (Gaussian density, Monte Carlo simulation); (v) inference and hypothesis testing; (vi) model evaluation; and (vii) parameter combinations.
The book is accompanied by the R package statforbiology and a dedicated website, both free to use. The website is regularly updated with new case studies and revisions to reflect ongoing developments in R.
Sommario
Chapter 1. Science, data, and experiments.- Chapter 2. The design of field experiments.- Chapter 3. Describing the observations.- Chapter 4. Cause-effect relationships.- Chapter 5. Stochastic models.- Chapter 6. A brief intro to statistical inference.- Chapter 7. Making Decisions under uncertainty.- Chapter 8. Checking fitted models.- Chapter 9. Linear/Nonlinear combinations of model parameters.- Chapter 10. Putting everything together.- Chapter 11. Extending regression models.- Chapter 12. A brief intro to mixed models.
Info autore
Andrea Onofri holds a PhD in "Crop Productivity" from the University of Perugia in Italy. Early in his career, he primarily focused on weed science, exploring weed-crop relationships, weed control methods, dose-response curves, and the ecotoxicological evaluation of herbicides. Later, for over 30 years, he has been extensively engaged in applying biostatistics to agriculture, concentrating on statistical model fitting, resampling techniques, time-to-event analyses, and genotype-by-environment interaction analyses. He has authored more than 80 papers in peer-reviewed journals and has developed several R packages. Currently, he serves as an Associate Professor at the University of Perugia, where he teaches multiple courses related to the subject of this book for master's and PhD students in the agricultural sciences.
Riassunto
Field research is a key element of scientific progress in agriculture and applied biology, bridging laboratory experiments in controlled conditions with on-farm trials conducted for demonstrative purposes. Like laboratory research, field studies must follow rigorous scientific protocols to yield reliable answers, while taking place in real-life settings to ensure practical relevance. The use of near-commercial-scale equipment and the need to address random variability, such as fertility gradients and pest invasions, have led to specific methodologies for experimental design and data analysis. These are seldom presented together in a concise, introductory format accessible to students and practitioners, and supported by real-world examples and reproducible code.
This book addresses that gap by focusing on small-plot field experiments evaluating genotypes, agronomic practices, pesticides, and plant protection methods. Using a 'learn-by-doing' approach, it presents selected examples and case studies, along with hands-on exercises. R is used as the primary tool for data analysis, with reproducible code snippets included. While not exhaustive, the book is tailored for a 6 ECTS introductory biometry course (approx. 54 hours), aimed at master’s and PhD students in agriculture and biology, and practitioners seeking foundational knowledge. No prior statistics background is required, though basic familiarity with R is assumed; an introductory appendix is provided for beginners.
Topics include: (i) experimental design and layout types; (ii) descriptive statistics; (iii) model fitting (ANOVA, regression, mixed models); (iv) stochastic models (Gaussian density, Monte Carlo simulation); (v) inference and hypothesis testing; (vi) model evaluation; and (vii) parameter combinations.
The book is accompanied by the R package ‘statforbiology’ and a dedicated website, both free to use. The website is regularly updated with new case studies and revisions to reflect ongoing developments in R.
