Create Regression Table for Multilevel Models — lmer

Produces a formatted table for multiple lmer fitted objects with various fit statistics and model comparisons.

Usage

lmer_table(
  models,
  fit_stats = c("fit", "random"),
  model_names = NULL,
  show_viewer = TRUE,
  silent = TRUE,
  digits = 2,
  ...
)

Arguments

models: List of fitted lmer objects
fit_stats: Character vector of statistics to include. Options: "ICC", "VIF", "REML", "fit", "LRT", "random", "random_p", "random_ci", "random_anova", "R2". See details.
model_names: Character vector of model names (default: auto-generated)
show_viewer: Logical. Show in RStudio viewer (TRUE) or go with stargazer's default (which is latex, see `type` argument in stargazer
silent: Logical. Suppress progress messages
digits: Integer. Number of digits for rounding
...: Additional arguments passed to stargazer

Value

Invisibly returns the HTML output path

Details

Fit statistics:

fit: Deviance, AIC, BIC, parameters, groups, observations, convergence
ICC: Intra-class correlation coefficient
R2: R-squared from empty model comparison
LRT: Likelihood ratio tests between nested models
random: Random effect variances
random_p: Random effect variances with bootstrapped significance sign
random_ci: Random effect variances with bootstrapped confidence intervals and sig sign
VIF: Maximum variance inflation factor
REML: Whether REML estimation was used

Examples

if (FALSE) { # \dontrun{
data("sleepstudy", package = "lme4")
m1 <- lmer(Reaction ~ Days + (1|Subject), sleepstudy)
m2 <- lmer(Reaction ~ Days + (1+Days|Subject), sleepstudy)
lmer_table(list(m1, m2))
} # }