“…The copyright holder for this preprint (which this version posted April 17, 2021. ; https://doi.org/10.1101/2021.04.11.439357 doi: bioRxiv preprint Gomes & Goerlitz, 2020;Zhao, Johnson-Bice & Roth, 2021). Although there is confusion over this recommendation, as some opt to use mixed-effects models despite this suggestion (Latta et al, 2018;Fugère, Lostchuck & Chapman, 2020;Gomes, Appel & Barber, 2020;Allen et al, 2021), likely because of the numerous advantages that mixed-effects models offer (Bolker, 2008;Kéry & Royle, 2015;Harrison et al, 2018), or fear of the consequences of pseudoreplication (although this can easily occur in mixed-effects models as well: Schielzeth & Forstmeier, 2009;Arnqvist, 2020). The trend to automatically follow this rule is likely exacerbated by the fact that authors or peer-reviewers can easily point out that this rule exists (Gelman & Hill, 2006;Harrison, 2015;Kéry & Royle, 2015;Harrison et al, 2018;Arnqvist, 2020), but may find it more difficult or time-consuming to make a nuanced argument against following such a rapidly growing rule.…”