Skip to contents

Advanced ANOVA analysis with comprehensive post hoc testing, effect sizes, and assumption checking. Addresses the critical issue where 68 percent of pathology studies fail proper multiple comparisons.

Usage

advancedanova(
  data,
  dependent,
  fixed,
  covariates,
  wls,
  model_type = "oneway",
  posthoc_method = "tukey",
  control_group = "",
  assumptions = TRUE,
  effect_sizes = TRUE,
  descriptives = TRUE,
  show_plots = TRUE,
  plot_type = "both",
  confidence_level = 0.95,
  alpha_level = 0.05,
  welch_correction = FALSE,
  robust_anova = FALSE
)

Arguments

data

the data as a data frame

dependent

The dependent variable from data, variable must be numeric

fixed

The grouping variable(s) from data, variable(s) must be a factor

covariates

Optional covariates for ANCOVA analysis

wls

Optional weights for weighted least squares

model_type

Type of ANOVA model to fit

posthoc_method

Post hoc comparison method

control_group

Control group for Dunnett's test comparisons

assumptions

Check and report ANOVA assumptions

effect_sizes

Calculate eta-squared, omega-squared, and Cohen's f

descriptives

Show descriptive statistics for each group

show_plots

Generate diagnostic and comparison plots

plot_type

Type of plots to generate

confidence_level

Confidence level for effect size confidence intervals

alpha_level

Alpha level for significance testing

welch_correction

Apply Welch correction for unequal variances

robust_anova

Use robust ANOVA methods for non-normal data

Value

A results object containing:

results$instructionsInstructions for using the Advanced ANOVA Suite
results$descriptivesDescriptive statistics for each group
results$assumptionsTests of ANOVA assumptions
results$anovaMain ANOVA results with effect sizes
results$tukeyTukey Honestly Significant Difference test results
results$gameshowellGames-Howell test for unequal variances
results$dunnettDunnett's test comparing treatments to control
results$bonferroniBonferroni-corrected pairwise comparisons
results$anovaplotViolin plots with boxplots and group means
results$diagnosticplotResidual plots and assumption checking
results$meansplotGroup means with confidence intervals
results$interpretationClinical context and interpretation guidelines

Tables can be converted to data frames with asDF or as.data.frame. For example:

results$descriptives$asDF

as.data.frame(results$descriptives)

Examples

# \donttest{
data('ToothGrowth')

advancedanova(data = ToothGrowth,
             dependent = len,
             fixed = supp)
#> Error in advancedanova(data = ToothGrowth, dependent = len, fixed = supp): argument "covariates" is missing, with no default
# }