Aarhus University Seal

Data project contributing to a research project 

Short description

The students are affiliated with a research project and work with a large data project using data science methods. Each group receives a project supervisor from the research group. The project work is supplemented by lectures and classroom instruction.

Motivation

The purpose of the course is to strengthen the students' communicative, analytical and technical skills through a number of projects in which an academic problem can be clarified and understood through a thorough analysis of related data. 

      

Learning objectives

The students make use of some of the competencies they have acquired in the first 18 months of their degree programme. The problems, analyses and results are presented and discussed in writing and orally by the students with a view to being able to: 

  • communicate a data analysis, 

  • argue and consider basic assumptions and prerequisites, 

  • derive and explain models and associated analyses, 

  • discuss the results of the analysis, 

  • consider which questions have been answered. 

Execution

PREPERATION

The educator's preperation

Divide the students into groups of 2-4 before the start of the course. Each group will be affiliated with a research project at Aarhus University in which data science is a central component. Each group is assigned a project supervisor, who is part of the research team on the research project. 

DURING THE SEMESTER

In plenum

Presentation by the research group

  • Representatives from the research group introduce data and describe the most important scientific questions and problems.

Academic presentations

  • During the semester, the students are taught mathematical, statistical and computer-science models, and the point of departure is a description of data as the outcome of a random process.

In groups

Choosing a topic

  • Together with the project supervisor, the students formulate a project that can improve and enlighten the supervisor's research.
  • Project examples: 
    • Causes of death in Denmark in a data-analytical perspective  
    • Improvement of a statistical method for monitoring plant diversity in Denmark  
    • Analysis of data from a large-scale study of the quality of life of cancer patients following radiation therapy 
    • Programming an app for visualising cardiopulmonary interactions during open chest surgery 

Project work

  • During the semester, the students analyse data from the research project using the models and methods presented in the teaching that they find appropriate for their project. They must defend their choice of methods for the presentation, analysis and interpretation of data.  

Giving and receiving feedback

  • During the course, the groups must give and receive oral and written feedback on presentations, written reports and materials. They must consider the academic problem, and critically discuss a data analysis. This is a requirement for taking the exam. 

Supervision

  • The project groups receive ongoing supervision from the assigned supervisor, who also participates as a representative from the research group. 

Written report

  • The students prepare and share their analyses on GitHub, and the results are gathered in a report.

THE EXAM

Individual

Oral exam

  • The exam consists of a 25-minute oral presentation of the group project and a discussion of the syllabus. 

RESOURCES FOR STUDENTS

SUPPORT FOR STUDENTS

  • The project groups receive ongoing supervision from the assigned supervisor, who also participates as a representative from the research group. 2 lessons per week for 14 weeks. 
     
  • Academic presentations by the research groups and the teacher. 2 hours of lectures per week for 14 weeks. 

Experiences

Outcomes 

The students developed competencies in communicating data analysis.

Challenges 

It can be challenging to find project supervisors with a relevant data set who have the time and desire to introduce students to their project. 


Basic information

Educator Asger Hobolth 
Faculty and department NAT, Department of Mathematics 
Degree programme Data Science 
Level of study Bachelor’s degree, 4th semester 
Course/subject Data project 
Number of students  20
Extent Whole course
Teaching format Lecture
Classroom instruction
Supervision
Implementation Spring 2023

Links and materials

Literature:  

  • Bayesian Statistical Methods (2019). Brian J. Reich and Sujit K. Ghosh. Taylor and Francis Group, Florida, USA. ISBN 978-0-815-37864-8 


Contact