Statnote 6: post-hoc ANOVA tests

Anthony Hilton, Richard A. Armstrong

Research output: Contribution to specialist publication or newspaperArticle


If data are analysed using ANOVA, and a significant F value obtained, a more detailed analysis of the differences between the treatment means will be required. The best option is to plan specific comparisons among the treatment means before the experiment is carried out and test them using ‘contrasts’. In some circumstances, post-hoc tests may be necessary and experimenters should think carefully which of the many tests available should be used. Different tests can lead to different conclusions and careful consideration as to the appropriate test should be given in each circumstance.
Original languageEnglish
Number of pages3
Specialist publicationMicrobiologist
Publication statusPublished - Sept 2006




Dive into the research topics of 'Statnote 6: post-hoc ANOVA tests'. Together they form a unique fingerprint.

Cite this