Biostatistik och experimentdesign, avancerad linjär modellering

The course is included as an optional course in the bachelor's program in biology, but is also given as independent course.

Access to the course requires completion of course BIO172, Biostatistics and experimental design, one-factor analyses, 7.5 credits

The course is a continuation of the course, BIO172, Biostatistics and experimental design, one-factor analyzes and focuses on advanced linear models. Consideration of experimental design is extended from BIO172. The course covers consideration of statistical interactions, model comparison and selection, generalized and mixed (hierarchical) models.

The course is rich in empirical examples from the lecturers' own laboratory and field work and includes several statistical software packages (eg R and SAS) and frameworks, including critiques of null hypothesis testing and alternative approaches to statistical analyses.

After completing the course, the student should be able to:

Knowledge and understanding

• Understand and explain the principles by which advanced linear models estimate coefficients

• Estimate and explain the meaning of partial relationships between predictors and response variables

• Understand the difference between and the relevance of random and fixed effects

Skills and ability

• Build advanced linear models with categorical and continuous predictors and interactions

• Integrate random effects into model construction appropriate to the data structure and scientific question

• Compare and choose among several alternative models using parsimony and information theory approaches

• Explain and defend model decisions used to address scientific questions

Evaluation ability and approach

• Critically evaluate reports of scientific findings by others and assess the weight of evidence for phenomena involving multiple variables

• Appreciate the ethical and practical implications of the reproducibility crisis in biostatistics

Teaching takes place through lectures, computer exercises, literature assignments and seminars with group work. Computer exercises are examined and literature assignments are discussed during seminars with compulsory attendance.

Language of instruction: English and Swedish The course is given in both Swedish and English. Please note that some literature and materials for exercises are only available in English.

The examination takes place in two stages. For the grade of G (Pass) correct submitted reports and participation in seminars and group discussions is required. For the grade VG (Pass with distinction) it is in addition to above requirements required that the student carries out a final written examination and there achieve 85% correct answers.

Occasion to supplement compulsory components can be given in the frame of the course or at the next course instance.

A student who has failed a test twice has the right to change examiner, unless weighty argument can be adduced. The application shall be sent to the department and has to be in writing.

The grading scale comprises: Pass with Distinction (VG), Pass (G) and Fail (U).

For the grade of G (Pass) correct submitted reports and participation in seminars andgroup discussions is required. For the grade VG (Pass with distinction) it is in additionto above requirements it is required that the student carries out in a final writtenexamination and there achieve 85% correct answers.

A written course evaluation is conducted at the end of the course. A summary of the course evaluation, along with a highlight of any changes, is presented for both the current and the upcoming course.