check_homogeneity: Check model for homogeneity of variances

In performance: Assessment of Regression Models Performance

Check model for homogeneity of variances

Description

Check model for homogeneity of variances between groups described by independent variables in a model.

Usage

check_homogeneity(x, method = c("bartlett", "fligner", "levene", "auto"), ...)

## S3 method for class 'afex_aov'
check_homogeneity(x, method = "levene", ...)

Arguments

x	A linear model or an ANOVA object.
method	Name of the method (underlying test) that should be performed to check the homogeneity of variances. May either be "levene" for Levene's Test for Homogeneity of Variance, "bartlett" for the Bartlett test (assuming normal distributed samples or groups), "fligner" for the Fligner-Killeen test (rank-based, non-parametric test), or "auto". In the latter case, Bartlett test is used if the model response is normal distributed, else Fligner-Killeen test is used.
...	Arguments passed down to car::leveneTest().

Value

Invisibly returns the p-value of the test statistics. A p-value < 0.05 indicates a significant difference in the variance between the groups.

Note

There is also a plot()-method implemented in the see-package.

See Also

Other functions to check model assumptions and and assess model quality: check_autocorrelation(), check_collinearity(), check_convergence(), check_heteroscedasticity(), check_model(), check_outliers(), check_overdispersion(), check_predictions(), check_singularity(), check_zeroinflation()

Examples

model <<- lm(len ~ supp + dose, data = ToothGrowth)
check_homogeneity(model)

# plot results
if (require("see")) {
  result <- check_homogeneity(model)
  plot(result)
}

performance documentation built on Nov. 2, 2023, 5:48 p.m.