Where there is no alternative to the use of animals in biomedical research, it is important that experiments are well designed and correctly analysed in order to minimise pain and maximize the chance of getting scientifically valid results. Experiments that use too few animals may fail to pick up biologically important effects, while those who use them incorrectly or wastefully may get invalid results while subjecting the animals to unnecessary pain, distress or lasting harm.
The Design of Animal Experiments is intended for all research scientists who use laboratory animals, with the aim of helping them to design their own experiments more effectively and/or to improve their ability to communicate with professional statisticians when necessary. It covers all randomised controlled experimental designs likely to be needed in laboratory animal research, with worked examples showing how they can be statistically analysed. It suggests the more widespread use of randomised block designs and shows how both males and females can be included in an experiment without the need to increase the total number of animals by using factorial designs. It also includes guidance on the choice of experimental animals. The book covers the learning outcomes of Module 10 and part (ii) of Module 11 of education and training under Directive 2010/63/EU.
Introduction and basic principles Choice of animals and their husbandry Understanding and controlling variability The analysis of variance Completely randomised (CR) single factor design Factorial experiments Randomised block (RB) designs Split plots, Latin squares, covariance and other techniques. Counts and proportions Regression and correlation The determination of sample size Seventeen steps in designing a randomised controlled animal experiment Concluding remarks Appendix 1. R and R Commander Appendix 2. Further reading Appendix 3. Learning outcomes for module 10 and part (ii) of Module 11 of the Education and training framework directive 2010/63/EU References Index