NNMK17004U Introduction to Ecological Data Analysis with R (REcoStat)
A good working knowledge of statistical data analysis and visualization is fundamental for most job functions within ecology, including conservation planning, environmental assessments or scientific research. It is also a necessary basis for being able to do analytical or field-based MSc thesis projects. This course aims at giving biology students the tools to perform independent data analysis for projects in ecology, and to understand and critically debate statistical data analysis from published reports and scientific papers. The main tool used in the course is the scientific programming language R, which is the de facto standard for ecological data analysis. The format mixes lectures and discussions with group exercises, and the students will work independently on data analysis projects to build the competence to do independent data analysis projects.
After completion of the course, the students are expected to be able to:
- Work independently to perform statistical analyses in ecology, including identifying scientific hypotheses and recognizing the statistical approach most suitable for testing them. This includes understanding the biological background and significance of different statistical tests and outcomes.
- Critically debate and replicate published analyses, both in published research papers and in reports addressing ecological questions of importance to nature management.
- Identify and acquire the necessary knowledge to conduct novel types of analysis.
use R to load data sets and do basic data analysis tasks
create custom functions and program simple simulations
use the R documentation to find solutions for coding problems
produce informative publication-quality figures, such as scatter plots, histograms and bar plots
test and summarize statistical models of ecological data
identify the assumptions of statistical tests and test if they are met
use standard linear regression, and derived techniques, such as spatial linear models, generalized linear models with different error families, phylogenetic regression, and random effects.
use the Rmarkdown syntax to produce a lab log of the analytical processes in a statistical analysis
- After completing the course, the students should be able to describe the basic elements of the R programming language and know the basic structure of academic programming languages.
- They should be familiar with the statistical methods available for analysis of observational data. In particular the students should be able to describe the issue of pseudoreplication and autocorrelation and detail the possible methods to deal with it.
- Finally, the students should know functions implemented in the R packages vegan for community ecological data analysis, nodiv and ape for working with macroecological data with phylogenetic trees, sp and raster for spatial data, ggplot2 for data visualization and dplyr for manipulation of data sets.
Handouts at the course
- Practical exercises
- Project work
- 7,5 ECTS
- Type of assessment
- Continuous assessmentAssesment based on 3-4 reports made during the course. The assesment is based on an overall assesment.
- Exam registration requirements
Active attendance during the course (80%).
- All aids allowed
- Marking scale
- passed/not passed
- Censorship form
- No external censorship
Several internal examiners
The student must hand in new or revised reports for the reexam.
If the requirements for participating in the exam is not fulfilled before the exam the student should hand in a report detailing an independent data analysis of an ecological dataset in R, presented in RMarkdown with figures and a discussion of key assumptions. The volume of the report should correspond to ~5 A4 pages. The report should be handed in no later than two weeks before the exam. This report may not be part of the exam.
Criteria for exam assesment
See learning outcomes.