Undervisning

Aim

The aim of the course is to

• obtain knowledge on existing health, social and welfare related registers
• design studies using existing registers
• identify epidemiological problems in register-based research

Content

The participants are introduced to registers and register-based studies, epidemiological problems within register-based research and how to design register-based studies.

Topics covered are:

• Registers and register-based studies,
• data management including data control,
• register-based epidemiology,
• monitoring development of health and disease in space and time,
• design and analysis of studies based on register data including cross-sectional, case-control, cohort and longitudinal studies.

The focus is on application of the concepts by practical exercises using SAS or STATA and working with own project.

Course description

The aim of the course is to

• create a data management plan for your PhD project
• give insights into how data management supports research projects
• understanding current principles and standards for documentation of quantitative and qualitative data
• provide considerations and introduce instruments to carry through data documentation and data control of research data from raw data to datasets for analysis
• understand and apply, in practice, the general rules of appropriate data management in accordance with responsible conduct of research

Data management plans are a key element of good data management and help you in the process of keeping track of the data that is part of a research project. A DMP is a formal document that outlines how to handle research data both during your research and after the research project is completed. It includes all parts of handling of primary materials and data throughout the research data lifecycle, which covers the data collection, organization, use, storage, contextualization, publication, preservation and sharing of research data.

The participants will through creating a datamanagement plan be introduced to central elements of datamanagement in research projects, workflows, relevant legislation, principles and standards for data documentation and data control, preparing quantitative and qualitative data for analysis and relevant legislation in relation to management and safekeeping of research data.

In the first three days, there will be lectures and exercises (full days) with frequent assignments related to the participant's own project. The following weeks participants (individually) work with their own project and prepare a course report (the data management plan of your PhD project). Beside conducting their own DMP the students will comment on a fellow course report. By the end of the third week the report is handed-in and at the seminar day (mandatory) each participant presents the DMP and the students will be opponents to each other.

Course description

The course will introduce statistical methods to reduce bias and explore causal relationships in modern epidemiology, with an emphasis on exercises conducted in either R or Stata. It will introduce the concept of the counterfactual approach for interpreting causality. Students will be taught techniques based on this approach to investigate mediating factors that contribute to causal relationships. Additionally, students will learn about Inverse-Probability-Weighting, G-computation, and Targeted maximum likelihood estimation as a means to reduce bias caused by confounding or selection. These technique offers several advantages compared to conventional methods that aim to control for systematic bias. The course will address causal considerations in time-to-event analyses and how pseudo-observations can be applied. The students will also gain an understanding of the importance of assessing interactions on both a multiplicative and additive scale and will be introduced to techniques to assess interactions on each scale. Finally, the course will delve into methodologies in quantitative bias analysis, such as introducing E- and G-Values. These tools are valuable for evaluating the robustness of associations against potential unmeasured or uncontrolled confounding.

Course description - Introduction to qualitative research methods

This 4-day PhD course is targeted PhD students at any stage of their study that may be entirely qualitative or include a qualitative component. The course is relevant for students who are at an introductory to intermediate level training on qualitative research.

The course aims for the student to gain theoretical and practical knowledge on how to analyze data from qualitative interviews and other types of unstructured, qualitative data. We aim to provide the student with knowledge and abilities to manage the different stages of creating a research design and the progression of a qualitative study from elaborating and conceptualizing the topic (research questions), deciding the data collection and analytic strategy and interpretation of results. 
Participants are supported in how to critically reflect on and strengthen methodological, theoretical and analytical approaches of their research study through lectures, student presentations, and discussions. 

Lectures in the morning will present qualitative research methods in relation to dominant philosophies of science and encourage the students to critically reflect on own research.
Workshops in the afternoon will include students presenting their own research focusing on conceptualization and methodological, theoretical and analytical perspectives according to the topic of the day.

Students are encouraged to participate actively, to debate their particular views, methodological problems, and research issues during class

Course description

Over the last five to ten years machine learning (ML) methods has gained widespread use with both sports and health data. ML methods can be used with both accelerometry or heart rate data for health or sports purposes or for simple clinical studies to find important patterns in the data. The possibilities seem almost endless. An important strength of the ML methods is that it can model highly complex data, which is common attribute of most sports and health data. However, the introduction of engines like the ChatGPT or Bard also suggests that understanding the strengths and weaknesses of this branch of statistical methods is important to disseminate quality health and physiological information from the sports and health data. 

Expected learning outcomes 

The present course will cover both unsupervised and supervised learning. Within unsupervised methods the focus is on principal component analysis and clustering, whereas with supervised methods we will cover both classification and numerical prediction using methods like decision trees and neural networks. The course will be highly practical, and you will get hands-on experience with model selection, learning, tuning and evaluating performance and generalizability. All work will be done using R. Prior experience in R is not required but would facilitate learning experience.