Vår 2017

HEL-8030 Applied Linear Regression Analysis - 3 stp

The course is administrated by

Institutt for samfunnsmedisin

Type of course

PhD Course. The course is available as a singular course.

Course overlap

MED-8003 Regression analysis 1 stp

Course contents

The course includes correlation as well as simple and multiple linear regression analysis, with application to medical and psychological areas. In particular, the following topics are being covered:

Model selection procedures
Interaction and confounding
Regression diagnostics
Transformation of variables
Fractional polynomial regression
Regression splines
Power analysis
Implementation of these techniques in statistical software packages

Application deadline

External applicants: Application deadline is 1 December for Spring semester and 1 June for Autumn semester. Application code 9303 in Søknadsweb.

If granted admission to the course you have to register for class and exam in Studentweb by 1 February in Sring semester and 1 September in autumn semester.

PhD students and students at Medical Student Research program at UiT The Arctic University of Norway go directly to Studentweb to register for class and exam by 1 February in Spring semester and 1 September in Autumn semester.

Admission requirements

PhD students, master students at a Medical Student Research Program, or holders of a Norwegian master´s degree of five years or 3+ 2 years (or equivalent) may be admitted. Valid documentation is a statement from your institution that you are a registered PhD student or master student at a Medical Student Research Program, or a Master´s Diploma with Diploma Supplement / English translation of the diploma. PhD students are exempt from semester fee.

For information about how to apply for admission, go to: http://uit.no/helsefak/forskning/phd/emner

Objective of the course

After taking the course, the students should

know about

the general linear model, its assumptions and applications
the concept of model selection and significance testing
how a power analysis can be usefully applied

be able to

identify different types of explanatory variables and correctly implement them in a linear regression model.
select the most appropriate model to study the relationship between one or more explanatory variables and one continuous response variable.
assess interaction and confounding variables.
interpret the results from linear regression models.
evaluate whether the assumptions of the regression models are fulfilled.
calculate power and sample size in linear models

have the general ability to

evaluate results from publications in medical/psychological journals where linear regression models are applied, and critically assess the validity of their use.

Language of instruction

English

Teaching methods

The program contains lectures, exercises with the use of PC, and review of the exercises using the software packages R, SPSS and STATA.

Assessment

Written assignment (10-20 pages) to be submitted. Evaluated as passed / failed.

The course is held annually. There will not be arranged a re-sit exam for this course.