Skip to content

This repository holds the code for an investigation how many groups you need in order to fit a random effect

Notifications You must be signed in to change notification settings

JohannesOberpriller/RandomEffect_Groups

Repository files navigation

Oberpriller, de Souza Leite, and Pichler, "Fixed or random? On the reliability of mixed-effects models for a small number of levels in grouping variables"

This subfolder contains the code to reproduce the results of Oberpriller, de Souza Leite, and Pichler, "Fixed or random? On the reliability of mixed-effects models for a small number of levels in grouping variables"

Simulations

We simulated two different mixed-effect model scenarios:

  1. Plant growth depends on Temperature Height ~ Temperature + (Temperature|Mountain range)

  2. Reproductive success depends on Temperature Reproductive success ~ Temperature + (Temperature|Mountain range)

Script to run simulations and fit models (5000 times for each number of mountain range):

scripts = list.files("Mountain_simulation/")
for(i in 1:length(scripts)) source(scripts[i])

Figures

Script to build Figures:

source("Figures_Main.R")
source("Figures_SI.R")
source("Figures_SI_glmmTMB.R")

Figures are saved in the "Figures" folder.

About

This repository holds the code for an investigation how many groups you need in order to fit a random effect

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages