Analysis of observational studies

Copenhagen Graduate School of Social Sciences

Department of Psychology, University of Copenhagen

Date and time: The seminar will take place three days, 9-11 August at 9:00-15:00

Successful participants of this course will have acquired profound knowledge on different ways of deriving scientific knowledge from observational studies, including the limitations, for example, with causal interpretation of effects.

Potential topics of the course include correlation studies, mediation analysis, structural equation modelling, propensity score matching, other matching procedures, case-control studies, register studies, causal inference, statistical methods for binary data and event times, Markov processes for multistage events, diagnostic studies, historical controls, nonrandomized studies, Bayes-statistics, STROBE-guideline.

PhD students of all fields can participate in this seminar. The minimum requirement is that students are interested in planning, conducting and analyzing observational research studies. Importantly, participants are asked to reflect on their use of methodology, their way to formulate research questions, and to conduct large projects. Requirements for passing the seminar are (a) attendance at each of the three days, (b) active participation, which will be evaluated based on classroom performance, (c) exercise sheets that are handed in.

Literature will be recommended and provided throughout the seminar.

Language: The entire seminar is held in English.

ECTS: 3

Course organiser and teacher: Professor Matthias Gondan-Rochon, Department of Psychology, Universität Innsbruck.

Max. numbers of participants: 20

Registration: Please register via the link in the box no later than 12 July 2021. Registration is closed, however there are still a few spots available. Please send an e-mail to the PhD administration if you are interest in a spot on the course.

Further information: For more information about the PhD course, please contact the PhD Administration (phd@hrsc.ku.dk).