Register      Login
Animal Production Science Animal Production Science Society
Food, fibre and pharmaceuticals from animals
RESEARCH ARTICLE (Open Access)

Perspective: are animal scientists forgetting the scientific method and the essential role of statistics?

J. L. Black A D , S. Diffey B and S. G. Nielsen C
+ Author Affiliations
- Author Affiliations

A John L Black Consulting, PO Box 4021, Warrimoo, NSW 2774, Australia.

B Centre for Crop and Disease Management, Curtin University, Bentley, NSW 6102, Australia.

C Research Office, Charles Sturt University, Wagga Wagga, NSW 2650, Australia.

D Corresponding author. Email: jblack@pnc.com.au

Animal Production Science 57(1) 16-19 https://doi.org/10.1071/AN15286
Submitted: 5 June 2015  Accepted: 21 August 2015   Published: 22 January 2016

Journal Compilation © CSIRO Publishing 2017 Open Access CC BY-NC-ND

Abstract

Animal scientists and their funding organisations need to ensure investment in research is maximised by strict adherence to the scientific method and the rigorous design and analysis of experiments. Statisticians should be considered as equals in the research process, engaged from the beginning of research projects and appropriately funded. The importance of experimental design that accounts for factors affecting the primary experiment measurement is illustrated in two examples. One shows how failure to involve a statistician at the beginning of a project resulted in considerable waste of resources. Subsequent engagement of professional statisticians with rigorous experimental design and analysis led to greatly increased precision in the standard error of an estimate for the digestible energy content of cereal grains for pigs from ± 0.35 MJ/kg to ± 0.16 MJ/kg. The other example shows the effect of the percentage of diets replicated during pelleting and of the total number of pigs required in the experiment on the P-values associated with detecting a pairwise difference between two grains differing in digestible energy content by 0.33 MJ/kg. Decisions based on these relationships have animal welfare and resource allocation implications.

Additional keywords: experimental design, measurement accuracy, treatment replications, digestible energy, pigs.


References

Baldwin RL (1995) ‘Modelling ruminant digestion and metabolism.’ (Chapman & Hall: London)

Black JL (2008) Premium Grains for Livestock Program: Component 1 – Coordination. Final Report. Grains R&D Corporation, Canberra, Australia. Available at http://ses.library.usyd.edu.au/handle/2123/1912 [Verified 5 November 2015]

Fisher RA (1938) Presidential address to the First Indian Statistical Congress. Sankhya 4, 14–17.

Kuehl RO (2000) ‘Design of experiments: statistical principles of research design and analysis.’ 2nd edn. (Duxbury/Thomson Learning Press: Pacific Grove, CA)

Levine TR (2013) A defence of publishing nonsignificant (ns) results. Communication Research Reports 30, 270–274.
A defence of publishing nonsignificant (ns) results.Crossref | GoogleScholarGoogle Scholar |

Smith AB, Lim P, Cullis BR (2006) The design and analysis of multi-phase plant breeding experiments. The Journal of Agricultural Science 144, 393–409.
The design and analysis of multi-phase plant breeding experiments.Crossref | GoogleScholarGoogle Scholar |