2018
DOI: 10.7287/peerj.preprints.3113
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

A brief introduction to mixed effects modelling and multi-model inference in ecology

Abstract: 31 32The use of linear mixed effects models (LMMs) is increasingly common in the analysis 33 of biological data. Whilst LMMs offer a flexible approach to modelling a broad range of 34 data types, ecological data are often complex and require complex model structures, 35 and the fitting and interpretation of such models is not always straightforward. The 36 ability to achieve robust biological inference requires that practitioners know how and 37 when to apply these tools. Here, we provide a general overview of… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

2
563
0
1

Year Published

2019
2019
2022
2022

Publication Types

Select...
10

Relationship

0
10

Authors

Journals

citations
Cited by 403 publications
(566 citation statements)
references
References 5 publications
2
563
0
1
Order By: Relevance
“…Full models are reported along with the simplified version of the models (i.e., reduced models; Crawley 2012). To compare models, we also used likelihood-ratio tests and report on changes in the likelihood ratio (LRT in mixed models) or F statistics (in generalized linear models) between two models of interest, and associated change in degrees of freedom ( df) and P values (Crawley 2012;Harrison et al 2018). Post hoc tests were performed using the multcomp package and glht function (Hothorn et al 2008).…”
Section: Resultsmentioning
confidence: 99%
“…Full models are reported along with the simplified version of the models (i.e., reduced models; Crawley 2012). To compare models, we also used likelihood-ratio tests and report on changes in the likelihood ratio (LRT in mixed models) or F statistics (in generalized linear models) between two models of interest, and associated change in degrees of freedom ( df) and P values (Crawley 2012;Harrison et al 2018). Post hoc tests were performed using the multcomp package and glht function (Hothorn et al 2008).…”
Section: Resultsmentioning
confidence: 99%
“…Post hoc analysis was performed with "multcomp" package and corrected for multiple testing. 28,29 We used the packages "ape" and "vegan" for MDS and PERMANOVA, respectively. 30 In this model, we accounted for repeated measurements using the "strata" argument.…”
Section: Resultsmentioning
confidence: 99%
“…We used AIC and Akaike model weights to reduce the whole set of models employing a dredging approach that retains a confidence subset of models that lay within six AIC units of the most informative model. This method removes models that have spurious parameter estimates due to poor model fit above the chosen AIC threshold (Richards 2005, Harrison et al 2018. The importance of each explanatory variable was judged according to AIC-weighted mean effect sizes averaged across the subset of regression models, and are presented as AIC-weighted slope estimates ±95% confidence intervals to estimate the significance of the effect of each predictor on body size (Table 1).…”
Section: Quantitative Analyses and Phylogenetic Controlmentioning
confidence: 99%