Challenging the value of a pseudoreplicated study: A comment to Liu et al. (2024)
: Kozlov, Mikhail V.
Publisher: ELSEVIER
: AMSTERDAM
: 2024
: South African Journal of Botany
: SOUTH AFRICAN JOURNAL OF BOTANY
: S AFR J BOT
: 174
: 180
: 182
: 3
: 0254-6299
: 1727-9321
DOI: https://doi.org/10.1016/j.sajb.2024.09.019
: https://doi.org/10.1016/j.sajb.2024.09.019
: https://research.utu.fi/converis/portal/detail/Publication/458221217
The paper by Liu et al. (2024) presents measurements of herbivory and six plant traits considered potential antiherbivore defences in seven plant species across two study plots: one mined and one adjacent unmined plot. I argue that this experimental design is inadequate for testing the authors' hypothesis that differences in herbivory and plant traits between these two plots are due to pollution impact, as the study lacks replication of pollution levels. This issue, which could have been easily addressed when designing the study, reflects insufficient attention to experimental design and seriously compromises the validity of the research findings. I briefly reiterate what pseudoreplication is and which fundamental principle of statistical inference it violates. (c) 2024 The Author(s). Published by Elsevier B.V. on behalf of SAAB. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/)
:
Research Council of Finland (grant 362731)
UTU Research Portal