Statistical Methods for the Environmental Research

1) Statistical basis — CFU 0.5 lectures + CFU 0.25 practical activities
Descriptive statistics and sample distributions. Overview of central tendency indices, with particular attention to mean and median, and variability indices, including sum of squares, mean square, standard deviation and coefficient of variation. Normal distribution and standardized normal distribution. Basic interpretation of data variability in environmental and agricultural datasets.

2) Basis of statistical inference — CFU 0.5 lectures + CFU 0.25 practical activities
Basic principles of statistical inference. Concepts of hypothesis testing, null and alternative hypotheses, significance level, p-value, type I and type II errors, and statistical power. Introduction to the interpretation of statistical tests in applied environmental research.

3) Analysis of variance and comparison among means — CFU 0.75 lectures + CFU 0.50 practical activities
One-way and two-way analysis of variance. Sources of variation, sums of squares, degrees of freedom, mean squares and F ratios. Interpretation of ANOVA tables. Multiple comparisons and interpretation of differences among treatments or groups. Application of ANOVA to environmental and agricultural datasets using statistical software.

4) Correlation, linear regression and multiple regression — CFU 0.75 lectures + CFU 0.50 practical activities
Simple linear regression and interpretation of intercept, regression coefficient, coefficient of determination and adjusted R². ANOVA of regression. Multiple regression and interpretation of the effects of explanatory variables. Variable selection, variable importance and evaluation of multicollinearity using the variance inflation factor. Application of regression models to datasets with continuous and/or categorical explanatory variables.

5) Linear models with continuous and categorical variables — CFU 0.50 lectures + CFU 0.25 practical activities
Use of linear models for the analysis of environmental and agricultural data. Combined use of continuous and discrete explanatory variables. Relationship between ANOVA, regression and general linear models. Interpretation of model outputs and methodological limits of the different approaches.

6) Introduction to multivariate analysis — CFU 0.50 lectures + CFU 0.25 practical activities
Basic principles of principal component analysis and cluster analysis. Standardization of variables. Interpretation of scores, loadings and biplot graphs. Distance measures, hierarchical clustering, k-means clustering and criteria for selecting the number of clusters, with particular attention to the silhouette graph.

7) Introduction to machine learning applied to environmental data — CFU 0.50 lectures + CFU 0.25 practical activities
Basic principles of machine learning. Training, validation and test datasets. Data splitting and prevention of data leakage. Identification of outliers, relevant variables, correlated variables and variables that introduce noise. Linear models as a baseline for machine learning approaches. Introduction to random forest and XGBoost, with basic concepts of hyperparameters and hyperparameter tuning.

8) Practical analytical workflow for environmental datasets — CFU 0.00 lectures + CFU 0.50 practical activities
Preparation of a structured analytical workflow for a dataset containing continuous and/or categorical independent variables and one dependent variable. Choice and justification of statistical and machine learning methods. Development of a concise written outline of the analysis and application of selected procedures using Python, or in a more limited way, Excel.

There are no differences between attending and non-attending students.