What you'll learn
- Develop a reliable and versatile data analysis framework
- Visualise your data with publication-ready figures
- Master general linear models: regression, ANOVA, etc.
- Refresh your statistical knowledge in a visual, intuitive way
- Learn and apply the principles of hypothesis testing and model selection
- Introduction to generalised linear models and to non-linear modelling
Requirements
- No programming experience needed. You'll learn how to use R from absolute 0.
- You should be familiar with statistical concepts covered in any introductory statistics course, such as: Normal distribution, model parameters, variance, standard deviation, standard error, F-test, p-value.
Description
This course is aimed at those that already have a theoretical understanding of statistical concepts and want to learn the practical side of data analysis.
Learning how to analyse data can be a daunting test. Applying the statistical knowledge learned from books to real-world scenarios can be challenging, and it's often made harder by seemingly complicated data analysis softwares.
This course will help you to develop a reliable data analysis pipeline, creating a solid basis that will make it easy for you to further your data analysis skills throughout your career.
We will use R, a free, state-of-the-art software environment for modelling, data handling, data analysis, and data visualisation.
We will start from installing R and taking baby steps to become familiar with the R programming language. We will then learn how to load data in R, how to visualise them with publication-level quality graphs, and how to analyse them.
I will provide you with the scripts that I use throughout the course, so that you can easily use them and adapt them to your own research objectives.
We will learn R one small step at a time, starting from absolute zero:
· how to enter data in R
· how to visualise data using function plot() and package ggplot2
· how to fit, interpret, and evaluate general linear models for a variety of study designs, including t test, ANOVA, regression, ANCOVA, and multiple regression scenarios
· how to fit polynomial regression
· an introduction to user-defined non-linear models
· an introduction to generalised linear models for non-normally distributed data (case study: count data)
· optimal data organisation and "data wrangling" - merging, subsetting, and summarising data
Who this course is for:
- Junior researchers moving their first steps into practical research in natural sciences, biology, medicine, etc.
Instructor
I am a quantitative ecologist (PhD) with a decade's worth of experience with performing and teaching data visualisation, data analysis, and modelling.
My expertise includes experimental design, general linear models, generalised linear models, mixed-effect models, non-linear models, multivariate statistics, and more. I do all my statistics, modelling, graphs, and data handling using the R software environment.
I have been involved in research on a variety of topics including population dynamics, community ecology, biodiversity, evolution and coevolution, biogeography, and spatial ecology, and my research has been published on international, peer-reviewed scientific journals. I like to translate the beauty of nature in quantitative terms, and to give data a voice.