Four-moment distributional analysis (Apidae)

Knowledge graph centered on Four-moment distributional analysis (Apidae) with 26 nodes and 108 connections. Top connected: Bombus, Ligusticum porteri, Bombus terrestris, Bombus appositus, Bombus impatiens.

Method synopsis

Bayesian mixed effects modeling of body size distributions using four statistical moments (mean, variance, skew, kurtosis) to characterize trait filtering processes. Uses robust measures and hierarchical modeling to account for species and site effects.

Synthesized from method descriptions across 2 papers using this protocol.

Procedure as described in the canonical source

Steps below were extracted from the paper that introduces this protocol — Intraspecific body size variation across distributional moments reveals trait filtering processes (2024), Journal of Animal Ecology. Implementations in other papers (listed below) may differ.

1. Calculate distributional moments
   Calculated four moments (mean, variance, skew, kurtosis) of body size distributions for each species-caste-site-year combination with sample sizes ≥30. Used robust coefficient of variation (RCV = IQR/median) for variance measure. Applied Fisher-Pearson standardized third moment coefficient for skewness and Pearson's measure for kurtosis.
   Quantities: Minimum 30 individuals per species-caste-site-year combination for calculationsDuration: Post-field season data processingConditions: Laboratory data analysis
   Equipment: R statistical software, moments package
2. Analyze statistical relationships
   Used Bayesian mixed effects models with brms package in R to estimate fixed effects of climatic and floral resource conditions on body size distributional moments. Modeled each moment independently with species as random effect and separate analyses for each caste. Ran three Markov chains with 4000 cycles each including 2000-iteration warmup phase.
   Quantities: Three Markov chains, 4000 cycles each, 2000-iteration warmupDuration: Computational analysis phaseConditions: R version 4.2.2 statistical computing environment
   Equipment: R software, brms package, Bayesian computation tools

Typical Equipment

• R software
• brms package
• moments package

Output Measurements

• distributional moments
• environmental effect estimates
• caste-specific responses
• trait filtering patterns

Papers Using This Protocol (2)

Species Studied With This Protocol