BIO415 Study Design, Data Analysis, and Reproducible Science

The course is connected to the following study programs

• Master's Programme in Coastal Ecology

Teaching language

English

Course contents

The course introduces students to study design and applied statistical methods commonly used in ecology, evolution, and environmental science. Students will learn how data can be organized, managed, analysed, and interpreted, through robust and repeatable workflows for data wrangling, visualisation, and analysis. The workflows will be implemented with the programming language R, the RStudio package, and the version control system Git.

The course will make students capable of generating, validating, and interpreting, basic linear models and give an introduction to extensions including generalised models, mixed models, and generalised additive models. The students will also be introduced to different approaches for model selection as well as methods for visualising and analysing multivariate data . The course provides insight on how to transform a scientific problem into statistical language, how to build a statistical model, how to analyse data, and how to validate results.

Every topic in the course is introduced and illustrated using examples. The students will be introduced to the essentials of programming and syntax-based data analysis using the open-source statistical and graphic environment R. The course will clarify best practices for reproducible and ethical research.

Learning outcomes

After successful completion of the course, students will

• be familiar with common challenges in data collection and data quality

• know basic principles and methods for collecting data and choose adequate experimental designs for different types of problems and situations

• be competent in detecting and describing major trends in data, by means of summarizing and graphical methods

• be capable of running linear models and generalized linear models (log-linear and logistic models) and interpreting the output from such models

• be familiar with more complex models, such as mixed- and nonlinear models and different methods for model selection

• be familiar with basic methods for visualising multivariate data and reducing dimensions in multivariate responses

 

After successful completion of the course, students will obtain the following skills using R, Rstudio and Git

• reshape and organise complex datasets

• build compelling data visualisations

• construct, and interpret the output of, linear, generalised, nonlinear, and hierarchical models

• explore and analyse data using methods for multivariate response

• use version control

 

Examination requirements

You need to have successfully completed the obligatory course components before being allowed to deliver the exam. More information about obligatory components is given in Canvas by the start of the course.

Teaching methods

The course is directed towards MSc students who need knowledge of statistics for their theses and will introduce examples from ecology, evolution, and environmental science. Teaching is focused on the practical application of statistical methods and implementation using R, RStudio, and Git. Classes will thus be a mixture of lectures and practical exercises, designed to build required skills to understand and perform analyses frequently encountered in the biological literature and for future independent work. Instruction will be given in English. More information will be given in Canvas at the start of the course. The estimated student workload in this course is 270 hours.

Evaluation

The study programme manager, in consultation with the student representative, decides the method of evaluation and whether the courses will have a midterm- or end of term evaluation, see also the Quality System, section 4.1. Information about evaluation method for the course will be posted on Canvas.

Admission for external candidates

No

Offered as Single Standing Module

Yes. Subject to availability or capacity.

Assessment methods and criteria

Portfolio. Graded assessment A-F. The portfolio consists of three parts each counting for one third of the final grade:

Data wrangling, quality control, and visualisation

Data analysis and reporting

Critical evaluation of results and visualisations

The portfolio will be delivered as a report consisting of R code and related output based on analysis of an individualised dataset. You need to get a passing grade for each part to be able to pass the exam.

There will not be arranged a postponed exam for the portfolio.