Experimental Design: the Good, the Bad and the Ugly
Wednesday, 06 March 2019, 9am to 5pm
Addressing some of the fundamental issues that are behind designing good (pre-clinical) experiments, the bedrock of reproducible science.
This course aims to address some of the fundamental issues that are behind designing good experiments (focussing on pre-clinical experiments exclusively), the bedrock of reproducible science.
The course will deal with those concepts, their links with statistical analyses generally and some of the traps that we have all fallen into. We will also address statistical issues relevant to animal research depending on the composition of the audience, in a session where the words “enjoyment” and “statistics” can hopefully share the same sentence.
A series of interactive lectures and scenarios delivered by Manuel Berdoy, BMS, Oxford.
At the end of the course, attendees should be able to:
- Describe some of the factors affecting reproducibility and external validity.
- List the different types of formal experimental designs (e.g. completely randomised, randomised block, repeated measures, Latin square and factorial experimental designs).
- Explain the concept of variability, its causes and methods of reducing it
- Describe possible causes of bias and ways of alleviating it
- Identify the experimental unit and recognise issues of non-independence (pseudo-replication).
- Describe the six factors affecting significance, including the meaning of statistical power and “p-values”.
- Identify formal ways of determining sample size.
- Explain the fundamental principles behind the output of an ANOVA, including “blocking” and “interactions”.