Unit EXPERIMENTAL METHODS IN AGRICULTURE
- Course
- Agricultural and environmental biotechnology
- Study-unit Code
- A002242
- Curriculum
- In all curricula
- Teacher
- Andrea Onofri
- Teachers
-
- Andrea Onofri
- Hours
- 54 ore - Andrea Onofri
- CFU
- 6
- Course Regulation
- Coorte 2021
- Offered
- 2021/22
- Learning activities
- Caratterizzante
- Area
- Discipline biotecnologiche agrarie
- Academic discipline
- AGR/02
- Type of study-unit
- Obbligatorio (Required)
- Type of learning activities
- Attività formativa monodisciplinare
- Language of instruction
- English
- Contents
- The course will be aimed to give students the theoretical background and practical tools to design scientifically sound experiments as well as proceed to correct analysis and presentation of results.
- Reference texts
- On-line e-book (linked on the UNISTUDIUM platform)
- Educational objectives
- Knowledge
1. Basic aspects on experimental design
2. Main experimental designs. When are they applied?
3. ANOVA: what it is and when it is applied.
4. How to check for the basic assumptions of ANOVa and what to do when they are not met
5. Linear regression: what it is and when it is applied.
6. Multiple Comparison Procedures. What they are and when they are used.
7. Non-Linear regression: what it is and when it is applied.
Practical skills
1. Design an experiment
2. Analyse the results of an experiment by using a statistical software
3. Check the basic assumptions for linear non-linear models.
4. Set up correction strategies
6. Build and fit basic nonlinear regression equations - Prerequisites
- Preliminary knowledge
1. Descriptive statistics: mean, mode, median, deviance, variance, standard deviation, coefficient of variability (essential)
2. Regression and correlation: fundamentals (important)
3. Population and sample, estimation process, estimation methods, confidence intervals (important)
4. Hypothesis testing, error rate, power and protection. Examples: t test and chi sqaure test (useful)
5. Fundamental skills with Excel: opening, closing and saving worksheets. Plotting data (bars and dispersion graphs). Entering functions (mean, sum etc...) (essential) - Teaching methods
- Lectures
Practicals
Audiovisual aids and study material:
Power Point slides for all lectures/practicals
Written assay for some lectures/practicals
Datasets - Other information
- The main infos are available on the UNISTUDIUM platform
- Learning verification modality
- Final oral exam with a practical evaluation, based on case study, to be analysed by using a personal computer
- Extended program
- The course is based on the fact that there are two main ways to obtain scientific information that is not already found in literature, i.e. (1) organise scientifically sound experiments and (2) appropriately use simulation models.
LECTURES (1.5 hours each, plus 45 min. of discussion, summary and presentation of case-studies)
1 - Measuring biological phenomena; variability of experimental data. Population, sample and sampling. Estimation methods and criteria. Statistical inference. Experimental units. Replication and pseudoreplication. Independence.
2 - Experimental design and ANOVA. Completely randomised designs (CR). Randomised complete block designs (RCB) and latin square designs. Factorial experiments. Examples.
3 - Split-plot, split-block and nested designs. Repeated experiments. Examples.
4 - ANOVA on CR and RCB designs. Examples.
5 - Problems with basic assumptions. Graphical analyses of residuals. Stabilising transformations. Examples.
6 - Multiple comparison testing. Examples.
7 - ANOVA on split-plot, split-block and repeated experiments designs. Examples.
8 - Regression and correlation. Polynomial regression. Statistical inference on regression analyses. Examples.
9 - Nonlinear regression analyses. Biologic assay. Examples.
10 - Nonlinear regression analyses. Degradation of xenobiotics. Crop growth curves. Other functions. Goodness and lack of fit. Examples.
PRACTICALS
Students will be exposed to some selected case studies and will be guided to their solution, by using the most appropriate statistical software.