Mixed Models Offer No Freedom from Degrees of Freedom
Abstract
Statistics matter greatly in biology, whether we like it or not. As a discipline with an empirical inclination, we are faced with data every day and we rely on inferential statistical models to make sense of it and to provide us with novel insights. Much of the time, the growing level of complexity and sophistication of the models we put to use in ecology and evolution have led to more appropriate analyses of our data. However, this is not always the case. Here, I draw attention to a classic flaw in inferential statistics that has resurfaced in a new flavor as a result of increased reliance on complex linear mixed models - the multifaceted and disturbingly persistent problem of pseudoreplication.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Problems with pseudoreplication due to issues with model convergence can lead to anticonservative p values and a high risk of false-positive claims.
“as well as problems with pseudoreplication due to issues with model convergence (Arnqvist, 2020; Barr et al., 2013; Forstmeier et al., 2017).”
From Violating the normality assumption may be the lesser of two evils (2021) · cites this paper as Arnqvist, 2020
When non-Gaussian models with multiple random effects fail to converge, researchers often settle for a model that ignores non-independence and yields estimates with inappropriately high confidence and statistical significance.
“which often lets researchers settle for a model that ignores non-independence and yields estimates with inappropriately high confidence and statistical significance (Arnqvist, 2020; Barr et al., 2013; Forstmeier et al., 2017)”
From Violating the normality assumption may be the lesser of two evils (2021) · cites this paper as Arnqvist, 2020
The citation lists haven’t been fetched yet
References 20 · Cited by 135
- Authors
- Arnqvist, G.
- Affiliations
- Uppsala University (SE)
- Corresponding author
- Arnqvist, Göran
- ORCID
- Arnqvist, Göran
- Venue
- Trends in Ecology & Evolution
- Publisher
- Elsevier BV
- Published
- 2020-04
- Volume
- 35
- Issue
- 4
- Pages
- 329-335
- ISSN
- 0169-5347 (ISSN-L)
- OpenAlex
- W3001636176
- Cited by
- 135
- References
- 20
- Citations by year
- 2026: 18 · 2025: 19 · 2024: 22 · 2023: 27 · 2022: 23 · 2021: 12 · 2020: 12 · 2018: 1
- FWCI
- 3.01
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Trends in Ecology & Evolution (publishedVersion)
- PubMed (publishedVersion)
- Funders
- Svenska Forskningsrådet Formas
- European Research Council
- Vetenskapsrådet
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Life Sciences › Agricultural and Biological Sciences › Ecology, Evolution, Behavior and Systematics › Agriculture, Soil, Plant Science
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Applications
- Keywords
- pseudoreplication, degrees of freedom, linear mixed models, statistical inference, model complexity, sample size
- MeSH terms
- Ecology, Research Design, Models, Statistical, Linear Models
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
