There are extensive sources of guidance on study design and statistical analysis from the EU Commission and many others. These, together with a statistician, should be consulted at an early stage. Direct oversight by a statistician is likely to be more effective than reliance on self-education from textbooks and websites, although many of these are excellent. An outline of the stages of the programme should be provided, indicating how each protocol will be used to achieve the objectives. Provision of a flow diagram or process map may be helpful. In addition to those elements described later in PREPARE, the following points should be considered:

1. A clear hypothesis and descriptions a priori of primary and secondary outcomes, to avoid HARKing.
2. Steps to minimise numbers and suffering of animals by appropriate statistical analysis, including the use of pilot studies.
3. Decisions on the power and significance levels to be used.
4. Definition of the experimental unit and number of animals in each unit
5. Choice of sample size and gender, age and/or developmental stage
6. Avoidance of bias, including “blinding” and randomisation (the procedure should be specified)
7. Inclusion and exclusion criteria

There are links to many more resources on experimental design and statistical analysis here.

Online protocols:

• Protocol Exchangefrom Nature.com
• protocols.io
• protocol-online.org
• Open Wetware

An overview of concerns in the literature about poor experimental design

Further reading

• Best practice for the design and statistical analysis of animal experiments (Palarea-Albaladejo & McKendrick, 2020)
• The reign of the p-value is over: what alternative analyses could we employ to fill the power vaccuum? (Halsey, 2019)
• Why most published research findings are false (John Ioannidis, 2005)
• Guidelines on statistics for researchers using laboratory animals: the essentials (Gosselin, 2018)
  Statistical experiment design for animal research (Sorzano & Parkinson, 2019, 317 pages)
• Statistical tests, P-values, confidence intervals, and power: a guide to misinterpretations (Greenland et al., 2016)
• Power to the People: Power, Negative Results and Sample Size (Gaskill & Garner, 2020)
• On determining sample size in experiments involving laboratory animals (M. Festing, 2018)
• How to decide your sample size when the power calculation is not straightforward (Bate, 2018)
• Sample size calculations: should the emperor's clothes be off the peg or made to measure? (Norman et al., 2012)
• What exactly is 'N' in cell culture and animal experiments? (Lazic et al., 2018)
• Author guidelines for displaying data, data analysis and statistical methods, from the American Society of Pharmacology and Experimental Therapeutics (ASPET)
• Two level factorial experiments (Smucker et al., 2019)
• Sex bias in preclinical research and an exploration of how to change the status quo (Karp & Reavey, 2018)
• Comparing phenotypic variation between inbred and outbred mice (Tuttle et al., 2018)
• The Weak Spots in Contemporary Science (and How to Fix Them) (Wickerts, 2017)
• Targeting next generations to change the common practice of underpowered research (Crutzen & Peters, 2017)
• Introducing Therioepistemology: the study of how knowledge is gained from animal research (Joe P. Garner et al.)
• How the Animal Welfare Body can help with reproducibility (Penny Hawkins)
• The development of response surface pathway design to reduce animal numbers in toxicity studies (Dewi et al., 2014)
• Threats to validity in the design and conduct of preclinical efficacy studies: a systematic review of guidelines for in vivo animal experiments (Henderson et al., 2013)
• Moving to a world beyond "p<0.05" (Wasserstein et al., 2019)
• Why null results do not mean no results: negative findings have implications for policy, practice and research (Miller-Halegoua, 2017)
• Rein in the four horsemen of irreproducibility (Bishop, 2019)