Variation: use it or misuse it – replication and its variants

Authors


  • This article is being published in The Journal of Physiology, Experimental Physiology, the British Journal of Pharmacology, Advances in Physiology Education, Microcirculation, and Clinical and Experimental Pharmacology and Physiology.

  • Gordon Drummond is Senior Statistics Editor for The Journal of Physiology.

  • Sarah Vowler is Senior Statistician in the Bioinformatics Core at Cancer Research UK's Cambridge Research Institute.

  • This article is the 9th in a series of articles on best practice in statistical reporting. All the articles can be found at http://jp.physoc.org/cgi/collection/stats_reporting

Email: g.b.drummond@ed.ac.uk

Key points

  • • Variation between measurements may be reduced if sampling is not random
  • • Fixed factors can reduce variation
  • • Replicates are repeated measurements from within an experimental unit
  • • Replication means that samples can no longer be considered random
  • • Comparisons of replicated values require correction, usually by reducing the effect of sample size
  • • Correction requires knowledge of the intra-class correlation

Ancillary