View in EDS

Species Overlap and Phylogenetic Relatedness Result in Community Statistical Non-Independence (and What to Do About It).

Saved in:
Bibliographic Details
Title: Species Overlap and Phylogenetic Relatedness Result in Community Statistical Non-Independence (and What to Do About It).
Authors: Tsang TPN; Department of Biological Sciences, University of Toronto-Scarborough, Toronto, Ontario, Canada., Cadotte MW; Department of Biological Sciences, University of Toronto-Scarborough, Toronto, Ontario, Canada.
Source: Ecology letters [Ecol Lett] 2025 Dec; Vol. 28 (12), pp. e70267.
Publication Type: Journal Article
Language: English
Journal Info: Publisher: Blackwell Publishing Country of Publication: England NLM ID: 101121949 Publication Model: Print Cited Medium: Internet ISSN: 1461-0248 (Electronic) Linking ISSN: 1461023X NLM ISO Abbreviation: Ecol Lett Subsets: MEDLINE
Imprint Name(s): Publication: Oxford, UK : Blackwell Publishing
Original Publication: Oxford, UK : [Paris, France] : Blackwell Science ; Centre national de la recherche scientifique, c1998-
MeSH Terms: Phylogeny* , Ecosystem* , Models, Biological* , Biodiversity*, Computer Simulation ; Linear Models ; Animals ; Biological Evolution
Abstract: Statistical autocorrelation between sampling units violates independence assumptions in many analyses. Here, we used simulations and empirical analyses to demonstrate how shared evolutionary history between species and species overlap among communities leads to an insidious form of autocorrelation, termed compositional autocorrelation. We simulated compositionally autocorrelated ecosystem functions across communities and assessed the type I error, statistical power and accuracy of slope estimates from naïve linear regression models and mixed models accounting for compositional autocorrelation. Mixed models maintained lower type I error, similar or higher statistical power, and more accurate slope estimates compared to linear regression. Re-analysing an empirical dataset, we found linear regression underestimated uncertainty in species richness effects for eight of 10 ecosystem functions. As species overlap and shared evolutionary history are common features in community data, our results highlight the need to explicitly consider compositional autocorrelation in statistical analyses to ensure correct inferences.
(© 2025 The Author(s). Ecology Letters published by John Wiley & Sons Ltd.)
References: J Anim Ecol. 2020 Jun;89(6):1482-1496. (PMID: 32163591)
Ecology. 2017 Dec;98(12):3175-3187. (PMID: 29023679)
Nature. 2006 Oct 26;443(7114):989-92. (PMID: 17066035)
Nature. 2025 Feb;638(8050):435-440. (PMID: 39779865)
Biol Rev Camb Philos Soc. 2019 Apr;94(2):648-661. (PMID: 30294844)
Proc Natl Acad Sci U S A. 2013 May 28;110(22):8996-9000. (PMID: 23674676)
Ecol Lett. 2005 May;8(5):548-57. (PMID: 21352459)
Nature. 2024 Apr;628(8009):788-794. (PMID: 38538788)
Trends Ecol Evol. 2018 Jul;33(7):549-564. (PMID: 29807839)
Ecology. 2023 Aug;104(8):e4104. (PMID: 37203421)
Ecol Lett. 2025 Aug;28(8):e70154. (PMID: 40772801)
Ecology. 2018 Feb;99(2):385-398. (PMID: 29121389)
Mol Ecol. 2023 May;32(10):2484-2503. (PMID: 35377502)
Am Nat. 2002 Dec;160(6):712-26. (PMID: 18707460)
Nature. 2001 Jul 5;412(6842):72-6. (PMID: 11452308)
Ecol Lett. 2017 Aug;20(8):989-996. (PMID: 28639274)
Nature. 2012 Jun 06;486(7401):59-67. (PMID: 22678280)
Ecol Lett. 2013 May;16 Suppl 1:116-27. (PMID: 23279851)
Ecol Lett. 2025 Dec;28(12):e70267. (PMID: 41319288)
Am J Hum Genet. 1973 Sep;25(5):471-92. (PMID: 4741844)
Nature. 1999 Oct 28;401(6756):877-84. (PMID: 10553904)
Ecology. 2011 May;92(5):1157-65. (PMID: 21661576)
Nat Commun. 2023 May 5;14(1):2607. (PMID: 37147282)
Ecol Lett. 2021 Dec;24(12):2537-2548. (PMID: 34532926)
Ecol Appl. 2016 Mar;26(2):448-62. (PMID: 27209787)
Biol Rev Camb Philos Soc. 2019 Aug;94(4):1220-1245. (PMID: 30724447)
Ecology. 2014 Jul;95(7):1717-23. (PMID: 25163105)
Ecology. 2016 Feb;97(2):286-93. (PMID: 27145604)
Ecol Lett. 2010 Feb;13(2):246-64. (PMID: 20102373)
Am Nat. 2020 Feb;195(2):145-165. (PMID: 32017624)
Ecology. 2011 Aug;92(8):1573-81. (PMID: 21905424)
Biol Rev Camb Philos Soc. 2007 Nov;82(4):591-605. (PMID: 17944619)
Ecol Appl. 2006 Feb;16(1):20-32. (PMID: 16705958)
Grant Information: #386151 Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada
Contributed Indexing: Keywords: autocorrelation; biodiversity; ecosystem function; phylogeny; species composition
Entry Date(s): Date Created: 20251130 Date Completed: 20251130 Latest Revision: 20251203
Update Code: 20251203
PubMed Central ID: PMC12665342
DOI: 10.1111/ele.70267
PMID: 41319288
Database: MEDLINE
Description
Abstract:Statistical autocorrelation between sampling units violates independence assumptions in many analyses. Here, we used simulations and empirical analyses to demonstrate how shared evolutionary history between species and species overlap among communities leads to an insidious form of autocorrelation, termed compositional autocorrelation. We simulated compositionally autocorrelated ecosystem functions across communities and assessed the type I error, statistical power and accuracy of slope estimates from naïve linear regression models and mixed models accounting for compositional autocorrelation. Mixed models maintained lower type I error, similar or higher statistical power, and more accurate slope estimates compared to linear regression. Re-analysing an empirical dataset, we found linear regression underestimated uncertainty in species richness effects for eight of 10 ecosystem functions. As species overlap and shared evolutionary history are common features in community data, our results highlight the need to explicitly consider compositional autocorrelation in statistical analyses to ensure correct inferences.<br /> (© 2025 The Author(s). Ecology Letters published by John Wiley & Sons Ltd.)
ISSN:1461-0248
DOI:10.1111/ele.70267