generalized linear mixed model (Sciuridae)

Knowledge graph centered on generalized linear mixed model (Sciuridae) with 72 nodes and 170 connections. Top connected: Marmota flaviventris, Rushes, mark-recapture (Sciuridae), small mammals, phenotypic plasticity.

Method synopsis

Bayesian statistical approach to simultaneously estimate individual lifetime means and variances in social traits from longitudinal data. Controls for age effects and repeated measures while quantifying individual-level variation.

Synthesized from method descriptions across 5 papers using this protocol.

Procedure from a recent peer-reviewed implementation

Steps below were extracted from the most recent peer-reviewed implementation of this protocol in the corpus — Conditional syndromes: Effect of human disturbance and age on the correlation between flight initiation distance and vigilance in marmots (2025), Behavioral Ecology and Sociobiology. The protocol was originally introduced by Locomotor ability and wariness in yellow-bellied marmots (2004), Ethology. Implementations in other papers (listed below) may differ.

1. Statistical analysis using bivariate mixed models
   Fitted bivariate mixed models using Bayesian MCMC methods (MCMCglmm in R) to estimate among-individual correlation between FID and vigilance. Used adults and yearlings but excluded juveniles. Log-transformed distance variables (FID, distance to burrow). Included disturbance level, age class, valley position, sex, substrate, and trial number as fixed effects. Year, colony, and individual ID included as random effects. Used parameter expanded priors with weak priors on variance parameters. Set residual correlation to zero to estimate only among-individual correlations.
   Quantities: 591 total individuals (227 with both FID and vigilance data, 336 vigilance only, 28 FID only), 5,100,000 MCMC iterations with 5,000 thinning and 100,000 burn-inDuration: Computational analysis phaseConditions: Analysis conducted on 18 years of data (2002-2020)
   Equipment: R software version 3.6.1, MCMCglmm package, Hadfield 2010 methods

Typical Equipment

• BRMS software
• R statistical software

Output Measurements

• lifetime trait means
• lifetime trait variances
• individual-level estimates

Papers Using This Protocol (5)

Species Studied With This Protocol