Get R_eady (10SMSTS-506/508)

Description

The course offers an introduction to data analysis in the transdisciplinary field of empirical (medical) research in the programming language R. The R system for statistical computing is openly available at https://www.r-project.org and provides a simple and flexible software environment for statistical analyses and graphics. Tailored to the application in empirical research, the course covers the basics of programming and data formats in R, as well as the essential steps of a data analysis including data manipulation, descriptive statistics, statistical tests and graphical representations. Reflections on research methodology and transdisciplinarity are addressed and critical thinking is encouraged.

The Get R_eady module enhances digital skills in the context of data analysis and graphics. It stimulates critical thinking and provides tools for reproducible research. In applications corresponding to the course participants' backgrounds, the concepts are explained. Hands-on sessions facilitate the understanding of the participants and enhance the discussion.

The course is building on problem based learning and held in an interactive and diverse format, with short lectures, demonstrations and hands on practical exercises to be solved in small groups.

Target group

Description

Prognostic models to predict future events have increasingly been used across different fields, e.g. in the medical sciences (clinical prediction models, personalized medicine, prognostic models), in legal data science (predictive analytics), political sciences (scientific prediction), or related.

The derivation and validation of such models poses specific challenges, that require knowledge of distinct methodological aspects in order to develop models that are internally valid and can be generalized out-of-sample. This course covers traditional statistical as well as machine learning approaches for model development, sample size calculation, variable selection, methodological outcomes for the assessment of model performance, as well as model validation. The course encourages critical thinking regarding published prognostic models’ validity across different fields of research.

The Get R_eady module enhances digital skills in the context of data analysis and graphics. It stimulates critical thinking and provides tools for reproducible research. In applications corresponding to the course participants' backgrounds, the concepts are explained. Hands-on sessions facilitate the understanding of the participants and enhance the discussion.

The course is building on problem based learning and held in an interactive and diverse format, with short lectures, demonstrations and hands on practical exercises to be solved in small groups.

Aims of the course 

• to equip participants with the essential tools to derive a prediction model, 
• to enable participants to apply and intepret suitable model diagnostics for different types of prediction models, 
• to empower participants to critically engage with and reflect on published prediction models

Target group

MA, PhD

ECTS Credits

1 ECTS

Course catalogue

You can find more information about the module here.