SFKM10168U Design and Evaluation of Animal Experiments

Volume 2013/2014
Education
Master's Programme in Laboratory Animal Science
Content
Classical experimental design: parallel groups, block design, cross-over design; Principles of design: factors, replication, randomization and blocking.
Quantitative and categorical data types. Statistical methods: analysis of variance and t-tests, non-parametric alternatives, chi-squared tests, analysis of covariance and factorial models. Dose-response models and regression. Repeated measurements and random effect factors.
Power and precision of an experiment.

Lectures will use the statisticak program R, but the participant may use his/her favorite statistical software if it does not entail severe limitations.
Learning Outcome
The course reviews classical types of experimental design together with planning issues like appropriate number of animals related to precision and power, randomization and applicable design types. Most commonly used statistical methods in animal experiments and their pitfalls are presented. Major emphasis is on possibilities of increasing precision via both design and statistical analysis with the aim of minimizing the number of animals needed.

At the end of the course the participant should have learned to

Knowledge
-describe classical types of experimental design used in animal experiments
- recognize types of experiments and data calling for advanced statistical methods

Skills
- plan and substantiate the design of an animal experiment, including issues of randomization, precision and power,
- analyse prototypes of such experiments statistically

Competences
- assess the use of conventional statistical methods for analysis of common types of animal experiments,
- interpret and report results of statistical analyses in accordance with statistical thinking
Literature
Festing MFW, Overend P, Das RG, Borja MC, Berdoy M: The Design of Animal experiments: reducing the use of animals in research through better experimental design, Laboratory Animals Handbooks No 14 2002

Morris, T.R. (1999). Experimental design and analysis in animal sciences. CABI Pulishing.

Both are highly recommended, but none of them required.

The first is more verbal and focused, the second is more statistical.
First period (from 4 weeks before gathering): e-learrning lectures and exercises based on (mostly) individual study. Second period (4 days gathered at SUND): lectures and discussions, exercises and minor projects. Third period (the subsequent month): individual work with exam project.
  • Category
  • Hours
  • Exam
  • 10
  • Lectures
  • 10
  • Preparation
  • 45
  • Project work
  • 25
  • Theory exercises
  • 20
  • Total
  • 110
Credit
4 ECTS
Type of assessment
Written examination
Individual report on a statistical project must be handed in one month after the week of teaching at SUND.
Weight: Assessment of report 100 %.
Aid
All aids allowed
Marking scale
7-point grading scale
Censorship form
No external censorship
More than one examiner
Criteria for exam assesment
The student can
-describe classical types of experimental design used in animal experiments
- recognize types of experiments and data calling for advanced statistical methods
- plan and substantiate the design of an animal experiment, including issues of randomization, precision and power,
- analyse prototypes of such experiments statistically
- assess the use of conventional statistical methods for analysis of common types of animal experiments,
- interpret and report results of statistical analyses in accordance with statistical thinking