Package 'autotestR'

autotestR

Description

designed to simplify the execution of the main statistical tests commonly used in the life sciences. It provides user-friendly functions that automatically generate plots and clear explanations, making statistical analysis more accessible for researchers and students.

Main features

• t test (independent and paired)

• Mann–Whitney test (Wilcoxon rank-sum)

• Multiple group comparison (t test or Mann–Whitney)

• Chi-squared test and Fisher’s exact test

• One-way ANOVA with Tukey HSD post hoc test

• Kruskal–Wallis test with Dunn post hoc test

• Pearson, Spearman, and Kendall correlation tests with automatic plots

• Diagnostic function that suggests the most appropriate statistical test

• Intuitive plots fully integrated into the functions

Basic usage

library(autotestR)

Independent t test

group1 <- rnorm(30, 10, 2) group2 <- rnorm(30, 12, 2) test.t(group1, group2)

Chi-squared test

var1 <- sample(c("A", "B"), 100, replace = TRUE) var2 <- sample(c("Yes", "No"), 100, replace = TRUE) test.chi(var1, var2)

Multiple test (t test or Mann–Whitney)

df <- data.frame( control = rnorm(30, 10), treatment = rnorm(30, 12), test1 = rnorm(30, 11), test2 = rnorm(30, 15) ) test.tmulti(df)

ANOVA with post hoc test

g1 <- rnorm(20, 5) g2 <- rnorm(20, 7) g3 <- rnorm(20, 6) test.anova(g1, g2, g3)

Correlation test

x <- rnorm(30) y <- x + rnorm(30, 0, 1) test.correlation(x, y)

Author(s)

Maintainer: Luiz Garcia [email protected] (ORCID)

---

pre.test() — Automatic statistical test suggestion

Description

Automatically identifies whether the input data are numeric or categorical and suggests the most appropriate statistical test.

Usage

pre.test(..., alpha = 0.05, help = FALSE, verbose = TRUE)

Arguments

...	Two or more vectors (numeric or categorical), or a data.frame with >= 2 columns
alpha	Significance level. Default = 0.05
help	Logical. If TRUE, shows detailed help
verbose	Logical. If TRUE, prints informative messages

Value

Invisible list with normality results, homogeneity or contingency table, and test recommendation

---

ANOVA test with automated assumption checking

Description

Performs ANOVA (and Tukey HSD) if data meet normality and homogeneity assumptions. Otherwise, automatically recommends Kruskal-Wallis/Dunn.

Usage

test.anova(
  ...,
  title = "ANOVA/Tukey HSD",
  xlab = "Group",
  ylab = "Value",
  style = c("boxplot", "violin", "mono", "halfeye"),
  help = FALSE,
  verbose = TRUE
)

Arguments

...	Vectors or a data.frame with >= 2 columns.
title	Plot title.
xlab	X-axis label.
ylab	Y-axis label.
style	Aesthetic style of the generated plot.
help	If TRUE, shows help.
verbose	If TRUE, shows detailed messages.

Value

An aov object or a recommendation message.

Examples

df <- data.frame(
  control = rnorm(30, 10),
  treatment = rnorm(30, 12),
  test = rnorm(30, 11)
)
test.anova(df)

---

Chi-square test

Description

Applies the Pearson chi-square test or Fisher's exact test to assess association between two categorical variables.

Usage

test.chi(
  x,
  y = NULL,
  title = "Chi-square Test",
  xlab = NULL,
  ylab = "Proportion",
  style = c("stacked", "barplot", "mosaic", "pie"),
  show_table = TRUE,
  help = FALSE,
  verbose = TRUE
)

Arguments

x	Categorical vector or data frame with two columns (group 1 and group 2).
y	Categorical vector (group 2). Required if x is a vector.
title	Plot title (string). Default: "Chi-square Test".
xlab	X-axis label in the plot (string). Default: NULL (uses variable name).
ylab	Y-axis label in the plot (string). Default: "Proportion".
style	Plot style generated by the function.
show_table	Logical. If TRUE, prints the contingency table to the console. Default: TRUE.
help	Logical. If TRUE, displays a detailed explanation of the function. Default: FALSE.
verbose	Logical. If TRUE, prints messages about the test and expected frequencies. Default: TRUE.

Value

Test result and contingency table.

Examples

data <- data.frame(
  control = c(rep("healthy", 50), rep("sick", 150)),
  treatment = c(rep("healthy", 100), rep("sick", 100))
)
test.chi(data)

---

Correlation Analysis with Automatic Method Selection

Description

Performs correlation analysis between two numeric variables using Pearson, Spearman, or Kendall methods. When method = "auto", the function automatically selects the most appropriate method based on normality, presence of ties, and proportion of outliers.

Usage

test.correlation(
  x,
  y = NULL,
  method = "auto",
  main = NULL,
  xlab = NULL,
  ylab = NULL,
  style = c("simple", "inference", "structure", "density", "distribution"),
  plot_normality = FALSE,
  help = FALSE,
  verbose = TRUE
)

## S3 method for class 'test.correlation'
print(x, ...)

Arguments

x	Numeric vector or data frame with exactly two numeric columns.
y	Numeric vector (optional if x is a data frame).
method	Correlation method: "auto" (default), "pearson", "spearman", or "kendall".
main	Plot title. If NULL, an automatic title is generated. If FALSE, no title is displayed.
xlab	Label for the x-axis. If NULL, a label is inferred from the input. If FALSE, no label is shown.
ylab	Label for the y-axis. If NULL, a label is inferred from the input. If FALSE, no label is shown.
style	Plot style: "simple" (default), "inference", "structure", "density", or "distribution".
plot_normality	Logical. If TRUE, displays QQ-plots for normality assessment.
help	Logical. If TRUE, displays a detailed usage guide.
verbose	Logical. If TRUE, prints information about method selection and diagnostics.
...	Additional arguments passed to other print methods (currently ignored)

Details

In addition to hypothesis testing, the function provides diagnostic information, effect size interpretation, and publication-ready visualizations.

Method selection in automatic mode follows these rules:

• If more than 5\ Kendall's tau is used.

• If both variables are normally distributed and no ties are present, Pearson's correlation is used.

• Otherwise, Spearman's rank correlation is applied.

Normality is assessed using Shapiro-Wilk (n <= 50), Anderson-Darling (50 < n <= 300), or Kolmogorov-Smirnov (n > 300) tests.

Value

An object of class "test.correlation" containing:

• call: Matched function call.

• data: Input data (if n <= 5000).

• n: Sample size.

• method: Selected correlation method.

• estimate: Correlation coefficient.

• effect: Effect size information (r, rho, or tau, and r² when applicable).

• conf.int: Confidence interval.

• p.value: P-value of the test.

• interpretation: Qualitative interpretation of effect size.

• diagnostics: Normality, ties, and outlier diagnostics.

• decision: Automatic selection rationale.

• htest: Underlying stats::cor.test object.

• plot: ggplot2 object.

See Also

cor.test, ggplot

Examples

# Pearson (approximately normal data)
set.seed(123)
x <- rnorm(50, sd = 0.1)
y <- x + rnorm(50, sd = 0.1)
test.correlation(x, y)

# Spearman (non-normal data)
set.seed(123)
x <- runif(300)
y <- log(x + 0.1) + rnorm(300, sd = 0.5)
test.correlation(x, y)

# Kendall (ties and outliers)
set.seed(123)
x <- runif(1000, 1, 100)
y <- sin(x) * 30 + rnorm(1000, 0, 10)
x[sample(1:500, 50)] <- 50
y[sample(1:500, 50)] <- 0
x_out <- runif(100, 10, 20)
y_out <- runif(100, 80, 120)
x <- c(x, x_out)
y <- c(y, y_out)
test.correlation(x, y)

---

Fisher's Exact Test

Description

Performs Fisher's Exact Test using two categorical vectors or a data frame with two columns, constructing a contingency table and optionally generating graphical visualizations.

Usage

test.fisher(
  x,
  y = NULL,
  title = "Fisher's Exact Test",
  xlab = NULL,
  ylab = "Proportion",
  style = c("stacked", "barplot", "mosaic", "pie"),
  show_table = TRUE,
  help = FALSE,
  verbose = TRUE
)

Arguments

x	Categorical vector or data frame with two columns.
y	Categorical vector (required if x_var is a vector).
title	Plot title (string). Default: "Fisher's Exact Test"
xlab	Name of the x-axis in the plot (string). Default: NULL (uses variable name)
ylab	Name of the y-axis in the plot (string). Default: "Proportion"
style	Plot style. Controls the visualization type
show_table	Logical. If TRUE, prints the contingency table to the console. Default: TRUE
help	Logical. If TRUE, shows detailed function explanation. Default: FALSE
verbose	Logical. If TRUE, prints detailed test messages. Default: TRUE

Value

Invisible object containing the Fisher test result.

Examples

data <- data.frame(control = c("healthy","healthy","sick","sick","sick"),
                   treatment = c("healthy","healthy","healthy","healthy","sick"))
test.fisher(data)

---

Test for interaction between a continuous predictor and a grouping variable

Description

Fits a linear model of the form y ~ x * by to evaluate whether the association between a continuous predictor and an outcome differs across groups. Optionally produces a publication-ready visualization of group-specific regression lines.

Usage

test.interaction(
  x,
  y,
  by,
  title = NULL,
  xlab = NULL,
  ylab = NULL,
  plot = TRUE,
  style = c("clean", "CI", "facet"),
  conf.level = 0.95,
  verbose = TRUE,
  help = FALSE
)

## S3 method for class 'test.interaction'
print(x, ...)

Arguments

x	Numeric vector representing the continuous predictor.
y	Numeric vector representing the continuous outcome.
by	Grouping variable defining the interaction. Must be coercible to a factor with at least two levels.
title	Optional title title for the plot.
xlab	Optional x-axis label.
ylab	Optional y-axis label.
plot	Logical. Should a plot be generated?
style	Plot style. One of "clean", "CI" or "facet".
conf.level	Confidence level for the interaction interval (default: 0.95).
verbose	Logical. If TRUE, prints detailed messages. Default: TRUE.
help	Logical. If TRUE, shows a detailed explanation of the function. Default: FALSE.
...	Additional arguments passed to other print methods (currently ignored)

Details

The interaction coefficient ($\beta$) represents the difference in regression slopes between groups, conditional on the reference level of by. The sign and magnitude of this coefficient depend on the chosen reference group.

Confidence intervals are emphasized as the primary inferential quantity.

Value

An object of class "test.interaction" containing:

• model: the fitted linear model,

• interaction: estimated interaction effects with confidence intervals,

• plot: a ggplot object (if plot = TRUE).

Examples

# Simple example: different trends between groups

set.seed(123)

n <- 60

marker <- rnorm(n, 10, 2)
group  <- rep(c("Control", "Treatment"), each = n/2)

# Same intercept, different slopes
response <- 2 +
  ifelse(group == "Control", 0.5, 1.2) * marker +
  rnorm(n, 0, 1)

test.interaction(marker, response, group)

---

Kruskal-Wallis Test with Dunn Post-hoc

Description

Performs the Kruskal-Wallis rank-sum test for comparing three or more independent groups, followed by Dunn's post-hoc test with multiple comparison adjustment.

Usage

test.kruskal(
  ...,
  title = "Kruskal-Wallis + Dunn",
  xlab = "Group",
  ylab = "Value",
  style = c("boxplot", "violin", "mono", "halfeye"),
  adjust = c("bonferroni", "holm", "BH"),
  help = FALSE,
  verbose = TRUE
)

Arguments

...	Numeric vectors representing groups, or a data frame with two or more columns (each column is treated as a group).
title	Character. Plot title.
xlab	Character. X-axis label.
ylab	Character. Y-axis label.
style	Character. Plot style. One of: "boxplot", "violin", "mono", or "halfeye".
adjust	Character. Method for p-value adjustment in Dunn's test. One of "bonferroni", "holm", or "BH".
help	Logical. If TRUE, displays a short help message and exits.
verbose	Logical. If TRUE, prints formatted statistical results to the console.

Details

This function is a non-parametric alternative to one-way ANOVA and is recommended when normality or homoscedasticity assumptions are violated.

Value

Invisibly returns a list with the following components:

type

Test type.

H

Kruskal-Wallis H statistic.

df

Degrees of freedom.

p

Global test p-value.

epsilon_sq

Epsilon-squared effect size.

epsilon_ci

Bootstrap confidence interval for effect size.

means_sd

Group means and standard deviations.

dunn

Dunn post-hoc results.

significant_pairs

Significant pairwise comparisons.

data

Long-format data used in the analysis.

Examples

set.seed(123)

n <- 25

df <- data.frame(
  control    = rexp(n, rate = 1),
  treatment1 = rexp(n, rate = 0.6),
  treatment2 = rgamma(n, shape = 2, scale = 1)
)

test.kruskal(df)

---

Student's t-test (with automatic diagnostics)

Description

Performs Student's t-test to compare means between two independent groups, with automatic checks for normality and homogeneity of variances. If assumptions are violated, the Mann-Whitney test is automatically applied (without generating a plot).

Usage

test.t(
  ...,
  title = TRUE,
  title_text = "t-test",
  xlab = "Group",
  ylab = "Value",
  style = c("boxplot", "violin", "mono", "halfeye"),
  help = FALSE,
  verbose = TRUE
)

Arguments

...	Two numeric vectors or a data frame with exactly two columns.
title	Logical. If true, return a plot entitled.
title_text	Plot title (string). Default: "t-test".
xlab	X-axis label in the plot (string). Default: "Group".
ylab	Y-axis label in the plot (string). Default: "Value".
style	Plot aesthetic generated by the function.
help	Logical. If TRUE, shows a detailed explanation of the function. Default: FALSE.
verbose	Logical. If TRUE, prints detailed messages. Default: TRUE.

Value

Invisible list with summary, test test_result, method and (optionally) plot.

Examples

set.seed(123)
df <- data.frame(
  control = rnorm(30, 10),
  treatment = rnorm(30, 15)
)
test.t(df)

---

Multiple pairwise comparisons with diagnostics

Description

Performs multiple pairwise comparisons using Student's t-test or Mann-Whitney test, with automatic diagnostics, effect sizes, confidence intervals, multiple testing correction and visualization.

Usage

test.tmulti(
  ...,
  comparisons = NULL,
  title = "Multiple comparisons (t / MW)",
  xlab = "",
  ylab = "Value",
  style = c("boxplot", "violin", "mono", "halfeye"),
  p_adjust = c("none", "holm", "BH", "bonferroni"),
  help = FALSE,
  verbose = TRUE
)

Arguments

...	Numeric vectors or a data.frame with groups in columns.
comparisons	List of character vectors specifying pairwise comparisons (e.g. list(c("A","B"), c("B","C"))). If NULL, all pairwise combinations are used.
title	Plot title.
xlab	X-axis label.
ylab	Y-axis label.
style	Plot style. One of "boxplot", "violin", "mono", or "halfeye".
p_adjust	Method for multiple testing correction. One of "none", "holm", "BH", "bonferroni".
help	Logical. If TRUE, prints usage examples.
verbose	Logical. If TRUE, prints results and plots.

Details

Normality is assessed using Shapiro-Wilk tests and homogeneity of variances using Levene's test. If assumptions are met, a pooled-variance t-test is used. Otherwise, the Mann-Whitney test is applied with bootstrap confidence intervals.

Effect sizes:

• Cohen's d for t-tests

• Rank-biserial correlation for Mann-Whitney

Value

A list with:

results

A tibble with test results.

plot

A ggplot object.

data_long

Long-format data used for plotting.

Examples

df <- data.frame(
  control   = rnorm(30, 10),
  treatment = rnorm(30, 12),
  test1     = rnorm(30, 11),
  test2     = rnorm(30, 15)
)

test.tmulti(
  df,
  comparisons = list(
    c("control", "treatment"),
    c("treatment", "test1")
  )
)

---

Paired t-test with advanced visualizations

Description

Performs a paired t-test between two numeric vectors (e.g., before vs after) or between two numeric columns of a data frame. Includes four visualization styles (boxplot, violin, mono, and half-eye).

Usage

test.tpaired(
  ...,
  title = "Paired t-test",
  xlab = "",
  ylab = "Value",
  style = c("boxplot", "violin", "mono", "halfeye"),
  connect = TRUE,
  help = FALSE,
  verbose = TRUE
)

Arguments

...	Two numeric vectors of equal length, or a data frame with exactly two numeric columns.
title	Plot title.
xlab	X-axis label.
ylab	Y-axis label.
style	Plot style: 1 Premium boxplot 2 Violin + minimal boxplot 3 mono 4 Half-eye (ggdist)
connect	Logical. If TRUE, connects paired observations.
help	If TRUE, displays detailed help.
verbose	If TRUE, prints progress messages.

Value

An invisible list containing:

summary_table

Group means and standard deviations

test_result

t-test result object (stats::t.test)

data

Data frame used for plotting

plot

ggplot2 object

Examples

before <- c(13, 12, 15, 14)
after  <- c(9, 11, 10, 10)
test.tpaired(before, after)

---

Mann-Whitney U Test

Description

Performs the Mann-Whitney (Wilcoxon rank-sum) test for comparing two independent groups, with statistical summary and graphical visualization.

Usage

test.u(
  ...,
  title = "Mann-Whitney Test",
  xlab = "Group",
  ylab = "Value",
  style = c("boxplot", "violin", "mono", "halfeye"),
  help = FALSE,
  verbose = TRUE
)

Arguments

...	Two numeric vectors or a data.frame with two numeric columns.
title	Plot title. Default: "Mann-Whitney Test".
xlab	Label for x-axis. Default: "Group".
ylab	Label for y-axis. Default: "Value".
style	Plot aesthetic style.
help	Logical. If TRUE, prints a detailed explanation. Default: FALSE.
verbose	Logical. If TRUE, prints detailed messages. Default: TRUE.

Value

Invisible list with:

summary

Group-wise statistical summary

test

Test result (htest object)

plot

ggplot2 visualization object

Examples

x <- c(1, 3, 5, 6)
y <- c(7, 8, 9, 12)
data <- data.frame(groupA = x, groupB = y)
test.u(data)

---