How to apply
For ph.d. candidates at USN: Register in Studentweb.
External applicants outside USN can use this form to apply for the course.
Deadline for applying is 15th of October.
The course requires a minimum of 5 registered participants to be held according to plan.
Content
This course provides students with a broad introduction to regression analyses with focus on linear regression, logistic regression, median regression, linear mixed effects models/repeated measures analysis.
The first part of the course provides essential knowledge in hypothesis testing and confidence intervals, sampling, and statistical inference. The second part is aimed at learning the basics of Stata including data management, data description, data manipulation, and data visualization.
Subsequently, the main part of the course will focus on regression analyses and their application to data using Stata. Detailed knowledge about prerequisites of the models, linearity, multicollinearity, confounding, effect modification, and variable selection will be given. We will consider how to choose among the different models: linear regression, logistic regression, median regression, and linear mixed effect models for a given dataset, and how to perform, implement and evaluate them in Stata. We will focus on the conceptual and practical understanding rather than on the mathematical formulas. Ideally, students should understand their data and the matching of design and research question, and then implement the model and interpret the results.
Learning outcome
Knowledge
- has basic knowledge in Stata, and are familiar with its structure and various interfaces
- has acquires expertise in performing linear regression, logistic regression, median regression and linear mixed effects models
- critical interpretation of multiple regression analysis
Skills
- can perform regression analysis in practice using software such as Stata
- can check that assumptions of the models are fulfilled
- can understand the concepts of confounding and effect modification (interactions)
- can select lvariables
- can analyse the results of designs that involve repeated or correlated measurements
General competences
- can understand and explain variation in the observed variables
- can apply analytical tools and critically evaluate one’s own and others’ research using similar analyses
- can justify the selection of appropriate methods on an individual basis
- can develop skills required to perform further studies on related topics