UV9218 – Linear Models

Course content

This course provides an introduction to principles, terminology, and strategies for statistical modelling with the linear model as initial framework for data analysis.

The linear model is a modelling workhorse for data analyses commonly referred to in the social and behavioural sciences as regression analysis, and is an essential building block towards more advanced regression-based model techniques such as multilevel analysis and structural equation modelling.

The emphasis in the course is on understanding the logic behind the modelling techniques and getting a hold of a proper non-naive interpretation of the model results.

The following topics are covered in class:

1. Simple regression: 1 predictor

2. Multiple regression: more predictors

3. Mini case studies

4. Model assumptions

5. Influential outliers

6. Categorical predictors

7. Second-order predictors: interaction

Although the course is in se platform/software independent, we will advance the use of the open-source statistical and graphic environment R during the computer labs.

UV9218 Linear Models is the PhD-level version of MAE4001 Linear Models, a compulsory course in the master's program, Assessment Measurement and Evaluation. The content, schedule and reading list for UV9218 Linear Models are the same as for MAE4001 Linear Models.

Learning outcome

Knowledge:

• Understand that models fit systematic data patterns but that residual random and/or systematic data patterns remain.
• Distinguish between a naïve experiment-inspired interpretation and a proper non-causal interpretation of model parameters.
• Understand that model-based inferences are affected by sampling variation and by the extent that model assumptions hold for the data.

Skills:

• Fit linear models to data in statistical software.
• Interpret model parameters and related statistics in light of the underlying data and study design.
• Hold a model-data dialogue using diagnostics to check the model and its inferential robustness.
• Write up the results of an analysis in an appropriate way.

Competence:

• Demonstrate a facility with linear modeling to answer well-defined research questions.
• Interpret published scientific research that uses these models and methods.
• Evaluate the tenability of associated inferences and knowledge claims.

Admission to the course

There is a limited number of seats due to joint teaching with the master’s level version of the course.

PhD candidates at the Faculty of Educational Sciences will be given priority, but it is also possible for others to apply for the course.

The deadline for registration is on the corresponding semester page for the course. 

Candidates admitted to a PhD-program at the Faculty of Educational Sciences (UV) can apply in StudentWeb.

Other applicants can apply by filling out and sending in a electronic registration form, which is found on the corresponding semester page for the course. 

Recommended previous knowledge

It is recommended to have had an introductory class covering descriptive (e.g., mean, variance, correlation) and inferential statistics (e.g., hypothesis tests) such as for instance MAE4000 Data Science.

Overlapping courses

• 3 credits overlap with UV9217 – Regression Analysis.
• 3 credits overlap with UV9214.
• 3 credits overlap with MAE4001 – Linear Models: Regression analysis.
• 1 credits overlap with UV9218U.
• 1 credits overlap with UV9217U.
• 1 credits overlap with UV9214U.

Teaching

This course combines lectures and computer labs with data analysis tasks in statistical software environments.

The course has joint teaching with the master course MAE4001 Linear Models.

Lectures are held by Associate Professor Björn Andersson. 

Obligatory course components:

• 80% attendance requirement for the lectures 
• Assignment must be passed

Schedule and literature: Please see the applicable semester page for the course. 

Examination

To obtain 3 credits, 80 % attendance, successful completion of the mandatory assignments and paper is required.

A more specific description of the mandatory assignments and paper will be given at the course.

Language of examination

The examination text is given in English, and you submit your response in English.

Grading scale

Grades are awarded on a pass/fail scale. Read more about the grading system.

More about examinations at UiO

• Use of sources and citations
• Special exam arrangements due to individual needs
• Withdrawal from an exam
• Illness at exams / postponed exams
• Explanation of grades and appeals
• Resitting an exam
• Cheating/attempted cheating

You will find further guides and resources at the web page on examinations at UiO.