Dealing with overdispersed count data in applied ecology

By S. Richards. 2007

648 citations recorded by the source corpus.

Subfields: Ecological Modeling, Nature and Landscape Conservation

View Paper: https://www.semanticscholar.org/paper/bdf84ffba99f1615f44cc099c3f1b0820d5fbd38

How researchers see this paper

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

In Ecological Modeling

In Nature and Landscape Conservation

Citations in the corpus

Model selection and model averaging in behavioural ecology: the utility of the IT-AIC framework (2010, methodology): “Thus, in this case, the NBD may be thought of as a model that implicitly encapsulates many unknown or unmeasured factors that contribute to an individual’s foraging success.”
Quarantine arthropod invasions in Europe: the role of climate, hosts and propagule pressure (2014, methodology): “Two of the four credible models were extensions of less complex and better-fitting models (Table 1) and were thus excluded from model averaging (Richards, 2008).”
When does invasive species removal lead to ecological recovery? Implications for management success (2018, methodology): “Models were also selected only if the AIC value of a complex model was less than the AIC value of the simpler model to avoid selecting overly complex models (Richards 2008).”
Using observation-level random effects to model overdispersion in count data in ecology and evolution (2014, background): “…Xavier A. Harrison, xav.harrison@gmail.com Academic editor Chiyuan Miao Additional Information and Declarations can be found on page 18 DOI 10.7717/peerj.616 Copyright 2014 Harrison Distributed under Creative Commons CC-BY 4.0 OPEN ACCESS Subjects Ecology, Evolutionary Studies, Statistics…”
When are eradication campaigns successful? A test of common assumptions (2012, methodology): “As a second rule, a model was only selected if its AIC value was less than the AIC value of all the simpler models within which it is nested in order to avoid selecting overly complex models (Richards 2008).”
Using information theory as a substitute for stepwise regression in ecology and behavior (2010, neutral): “In contrast to critical P values, it is poorly known whether arbitrary threshold differences in IT criteria that are currently applied are too permissive or too restrictive, because there is simply insufficient experience to tell this (Richards 2005, 2008; Ward 2008).”
Ecologists overestimate the importance of predictor variables in model averaging: a plea for cautious interpretations (2014, methodology): “Methods in Ecology and Evolution © 2014 British Ecological Society, Methods in Ecology and Evolution, 5, 983–991 set of models built by removing nested models complicating the first-rankedmodel within the subset ofmodels withDi 6 (Richards 2008, Richards, Whittingham & Stephens 2011).”
The consequences of checking for zero‐inflation and overdispersion in the analysis of count data (2019, methodology): “they do not; see Zuur et al. (2010). In the case of overdispersion, popular alternatives to the Poison GLM include the Quasi-Poisson (QP) model (Wedderburn 1974) and the Negative Binomial (NB) model (Richards 2008, Linden & Mantyniemi 2011). Note that when selecting between the QP and NB models,”
Multimodel inference in ecology and evolution: challenges and solutions (2011, methodology): “Simulation studies have shown that in certain circumstances, choosing the ‘best’ model (based on AICC for example) may provide similar parameter estimates when compared to model averaging.”
Environmental DNA surveillance for invertebrate species: advantages and technical limitations to detect invasive crayfish Procambarus clarkii in freshwater ponds. (2014, methodology): “As suggested by Richards (2008), a model was selected only if ΔAICc ≤ 6, and its AICc was less than the AICc of all the simpler models within which it was nested.”
Living in a landscape of fear: the impact of predation, resource availability and habitat structure on primate range use (2014, methodology): “Models were selected based on the AIC selection method suggested by Richards (2008).”
A brief guide to model selection, multimodel inference and model averaging in behavioural ecology using Akaike’s information criterion (2010, background): “The causes of overdispersion are numerous, and there are several ways of dealing with it (see Richards 2008 for a recent primer).”
Repeatability for Gaussian and non‐Gaussian data: a practical guide for biologists (2010, background): “Notably, some count data are better described by negative binomial distributions when overdispersion is large and such count data are fairly common (Richards, 2008).”