© Saxon. 2015
Published: 23 October 2015
Oat plants grown at an agricultural research facility produce higher yields in Field 1 than in Field 2, under well fertilised conditions and with similar weather exposure; all oat plants in both fields are healthy and show no sign of disease. In this study, the authors hypothesised that the soil microbial community might be different in each field, and these differences might explain the difference in oat plant growth. They carried out a metagenomic analysis of the 16 s ribosomal ‘signature’ sequences from bacteria in 50 randomly located soil samples in each field to determine the composition of the bacterial community. The study identified >1000 species, most of which were present in both fields. The authors identified two plant growth-promoting species that were significantly reduced in soil from Field 2 (Student’s t-test P < 0.05), and concluded that these species might have contributed to reduced yield.
However, significance at the level of P < 0.05 in a Student’s t-test means that the false positive rate — incorrectly rejecting the null hypothesis that there is no difference between the groups — is 5 %. When the number of comparisons is large, the likelihood of false positive results is greater: for the 1000 comparisons in this study, around 50 are expected falsely to show a significant increase or decrease in level. The researchers in this case did not correct for multiple comparisons, but several methods exist for doing so. The simplest of these is the Bonferroni correction, which uses a P value calculated as 0.05/n, where n is the number of comparisons made. But this method can ‘over-correct’, leading to significant differences being overlooked; Nakagawa’s 2004 review provides some more detail on this subject . More complex methods, such as the Benjamini-Hochberg procedure, allow for a slightly less strict correction, reducing the rate of false negative results that occur with the Bonferroni method . This is particularly useful when the number of comparisons is very large, as is the case in this study.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Nakagawa S. A farewell to Bonferroni: the problems of low statistical power and publication bias. Behav Ecol. 2004;15:1044–5.View ArticleGoogle Scholar
- Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J Roy Stat Soc B Methodol. 1995;57:289–300.Google Scholar