The mistreatment of covariate interaction terms in linear model analyses of behavioural and evolutionary ecology studies

By Leif Engqvist. 2005

761 citations recorded by the source corpus.

Subfields: Ecology, Evolution, Behavior and Systematics, Nature and Landscape Conservation

View Paper: https://www.semanticscholar.org/paper/b02643659a70b66dffd97aa1eb6a87af6b7d5248

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Ecology, Evolution, Behavior and Systematics

In Nature and Landscape Conservation

Citations in the corpus

Spatio-temporal heterogeneity in abiotic factors modulate multiple ontogenetic shifts between competition and facilitation (2010, background): “…and these differences could mask the effect of an experimental treatment, its inclusion as a covariate has been recommended to gain power when testing the effects of the factors of interest, particularly when a high intra-group variability is found (see Engqvist, 2005 and references therein).”
Quantifying individual variation in behaviour: mixed-effect modelling approaches. (2013, background): “…of the sorts of questions that can be addressed with MMs and variance decomposition to avoid misunderstanding (e.g. Bennington & Thayne 1994; Engqvist 2005) of what information these methods can provide (Table 1 details where these questions are addressed): 1 Do individuals differ in…”
Diet quality affects an attractive white plumage pattern in dark-eyed juncos (Junco hyemalis) (2007, background): “However, brighter white plumage has been linked to dominance, attractiveness, and mating success in black-capped chickadees (Poecile atricapillus (Doucet et al. 2005; Mennill et al. 2003; Woodcock et al. 2005), so it is reasonable to expect that tail-white brightness may also be important in juncos.”
Simple means to improve the interpretability of regression coefficients (2010, background): “There are well-justified warnings against the interpretation of main effects when the input variables are involved in interactions (Aiken & West 1991; Engqvist 2005; Gelman & Hill 2007).”
Conflictual student–teacher relationship, emotional and behavioral problems, prosocial behavior, and their associations with bullies, victims, and bullies/victims (2018, methodology): “As a final step, we ran the model again without including non-significant interaction effects (Engqvist, 2005).”
Multiple Male Reproductive Morphs in the Invasive Round Goby (Apollonia melanostoma) (2009, background): “05), they were removed from the model (Engqvist, 2005) and only the subsequent model F values were reported.”
Nested by design: model fitting and interpretation in a mixed model era (2013, background): “This point relates to a well-known issue in model fitting, namely the interpretation of interaction effects (Aiken &West 1991; Engqvist 2005).”
Assessing translocation outcome: Comparing behavioral and physiological aspects of translocated and resident African elephants (Loxodonta africana) (2009, positive): “Since no interactions among the effects of the model were statistically significant, they were not included in the final model (Engqvist , 2005).”
Using information theory as a substitute for stepwise regression in ecology and behavior (2010, neutral): “In the presence of weak interaction terms, the main effect estimates will have large standard errors, and they will also be biased unless very restrictive conditions are met (Aiken and West 1991; Engqvist 2005).”