A versatile workflow for linear modelling in R

Abstract

Linear models are applied widely to analyse empirical data. Modern software allows implementation of linear models with a few clicks or lines of code. While convenient, this increases the risk of ignoring essential assessment steps. Indeed, inappropriate application of linear models is an important source of inaccurate statistical inference. Despite extensive guidance and detailed demonstration of exemplary analyses, many users struggle to implement and assess their own models. To fill this gap, we present a versatile R-workflow template that facilitates (Generalized) Linear (Mixed) Model analyses. The script guides users from data exploration through model formulation, assessment and refinement to the graphical and numerical presentation of results. The workflow accommodates a variety of data types, distribution families, and dependency structures that arise from hierarchical sampling. To apply the routine, minimal coding skills are required for data preparation, naming of variables of interest, linear model formulation, and settings for summary graphs. Beyond that, default functions are provided for visual data exploration and model assessment. Focused on graphs, model assessment offers qualitative feedback and guidance on model refinement, pointing to more detailed or advanced literature where appropriate. With this workflow, we hope to contribute to research transparency, comparability, and reproducibility.

Scorpionfish adjust skin pattern contrast on different backgrounds

“…We implemented generalised linear mixed models with the glmmTMB R-package (Brooks et al, 2017) following a custom-written guided linear modelling R-routine (Santon et al, 2023). Model assessment followed the guidance of Santon et al (2023), focusing on the inspection of the distribution of randomised quantile residuals, computed with the R-package DHARMa (Hartig, 2022), within and among factor predictor levels that were included or not in the models, and performed posterior predictive checks to assess model dispersion and overall model fit. Models were initially implemented using the most appropriate family distribution for the nature of the response variable.…”

“…We implemented generalised linear mixed models with the glmmTMB R‐package (Brooks et al., 2017) following a custom‐written guided linear modelling R‐routine (Santon et al., 2023). Model assessment followed the guidance of Santon et al.…”

Habitat selection and refuge‐use by a color polymorphic salamander reveal behavioral niche differences

“…For analyses, we used the R‐workflow template (and packages therein) for generalized linear mixed model analysis (glmmTMB) provided by Santon et al. ( 2023 ). The template was used to select the most appropriate models by evaluating residual dispersion, zero inflation, and the overall model fits provided by different residual distribution families and link functions.…”

My friend MIROSLAV: A Hackable Open-Source Hardware and Software Platform for High-Throughput Rodent Activity Monitoring in the Home Cage

“…Estimated marginal means of the parameters are produced using the emmeans package [27]. The data and model specification were checked against the workflow for linear modelling in R by Santon et al [28].…”