The Experimental Design Assistant - EDA

The Experimental Design Assistant (EDA) is an online tool to guide researchers through the design of their experiments, helping to ensure that they use the minimum number of animals consistent with their scientific objectives, methods to reduce subjective bias, and appropriate statistical analysis.

System requirements

 

We recommend using the EDA with the latest stable release of Chrome. Alternatively, the latest stable release of Mozilla Firefox or Safari can also be used.

 

The EDA is not currently compatible with mobiles and tablets. We intend to support IE11 and Edge in later releases. Please note that the EDA will not load in Internet Explorer browsers prior to IE11. 

 

Using the EDA in an unsupported or out-of-date browser may result in:

  • being unable to access the EDA
  • loss of functionality of EDA features
  • display issues while using the designer

Background to the Experimental Design Assistant

From the blog: Improving the design of animal experiments: Introducing the Experimental Design Assistant (EDA).

In 2009, we published a survey of the peer-reviewed literature which assessed the quality of publically funded animal research in the US and the UK. The survey identified areas for improvement in:

  • Experimental design
  • Statistical analysis
  • Reporting of studies

In response to the survey we produced guidelines for the reporting of animal experiments – the ARRIVE guidelines, which were published in 2010 and subsequently endorsed by many journals, research funders, universities and learned societies.

To build on this work and provide further support to improve the quality of animal experiments, we developed the EDA which was launched in 2015. The EDA is developed in collaboration with an expert working group and Certus Technology, a company which specialises in innovative software for the life sciences.

How does the EDA further the implementation of the 3Rs?

Good experimental design can minimise animal use in two ways:

  • By accounting for the influence of variables and addressing sources of bias, an adequately designed experiment will yield robust and reproducible data, ensuring that the data from every animal is utilised to its full potential.
  • An efficient use of statistics can reduce the number of animals required and maximise the information obtained per experiment. More complex designs for example can help researchers identify factors which influence the experimental results, providing more information about the model they are using.

Working group membership

Professor Clare Stanford (Chair) University College London
Dr Simon Bate GlaxoSmithKline
Dr Manuel Berdoy University of Oxford
Dr Robin Clark Envigo
Professor Innes Cuthill University of Bristol
Dr Derek Fry University of Manchester
Dr Natasha Karp AstraZeneca
Professor Malcolm Macleod University of Edinburgh
Dr Lawrence Moon King’s College London
Dr Richard Preziosi University of Manchester

 

Related content

Back to top