EasyStat: Automated Statistical Analysis, Visualization and Multi-Format Narrative Reporting

Description

EasyStat provides a unified pipeline for automated statistical analysis and plain-language narrative reporting in R. The four-step pipeline — Core Statistical Engine, Metric Extractor (broom), Narrative Generator Module, and Unified Result Object — transforms raw data into publication-ready output with a single function call.

Descriptive statistics: easy_describe, easy_group_summary

Inferential tests: easy_regression, easy_ttest, easy_anova, easy_chisq, easy_ztest, easy_ftest, easy_correlation

Visualizations: easy_histogram, easy_boxplot, easy_scatter, easy_barplot, easy_qqplot, easy_density, easy_correlation_heatmap, easy_autoplot

Reporting: export_to_word, theme_easystat

Authors

Mahesh Divakaran (Amity School of Applied Sciences, Amity University Lucknow), Gunjan Singh (Amity School of Applied Sciences, Amity University Lucknow), Jayadevan Shreedharan (Gulf Medical University)

Author(s)

Maintainer: Mahesh Divakaran imaheshdivakaran@gmail.com (Amity School of Applied Sciences, Amity University Lucknow)

Authors:

• Gunjan Singh (Amity School of Applied Sciences, Amity University Lucknow)

• Jayadevan Shreedharan (Gulf Medical University)

See Also

Useful links:

• https://github.com/itsmdivakaran/Easystat

• https://itsmdivakaran.github.io/Easystat/

• Report bugs at https://github.com/itsmdivakaran/Easystat/issues

Run a One-Way ANOVA with Automated Narrative Reporting

Description

Executes a one-way ANOVA using stats::aov(), extracts key metrics via broom, computes eta-squared as an effect-size measure, and generates a plain-language narrative via the Narrative Generator Module.

Usage

easy_anova(formula, data, alpha = 0.05)

Arguments

Value

An object of class "easystat_result" with:

test_type

Character: "anova"

formula_str

Character string of the formula used

raw_model

The raw aov object

coefficients_table

ANOVA table (SS, df, MS, F, p)

model_fit_table

Summary metrics (F-statistic, eta-squared, p-value)

explanation

Plain-language narrative string

Examples

result <- easy_anova(Sepal.Length ~ Species, data = iris)
print(result)

Automatically Plot an EasyStat Result

Description

Chooses the most appropriate plot type based on the test_type of an easystat_result object and renders it.

Usage

easy_autoplot(result, data = NULL, ...)

Arguments

Value

An "easystat_result" plot object, invisibly.

Annotated Bar Chart

Description

Creates a frequency bar chart for categorical variables, or a mean-and-error bar chart for numeric outcomes grouped by a factor.

Usage

easy_barplot(
  x,
  data,
  group_by = NULL,
  stat = c("count", "mean"),
  fill_palette = NULL,
  title = NULL
)

Arguments

Value

An "easystat_result" object with plot_object.

Grouped Boxplot with Outlier Detection

Description

Produces a boxplot for one variable, optionally grouped by a factor. Adds a jittered dot overlay, labels each group's median, and highlights outliers.

Usage

easy_boxplot(formula, data, fill_palette = NULL, notch = FALSE, title = NULL)

Arguments

Value

An "easystat_result" object with plot_object.

Chi-Square Tests with Automated Narrative Reporting

Description

Runs either a chi-square test of independence (two categorical variables) or a goodness-of-fit test (one variable vs. expected proportions), extracts Cramér's V as the effect-size measure, and generates a plain-language narrative via the Narrative Generator Module.

Usage

easy_chisq(x, y = NULL, data = NULL, p = NULL, correct = TRUE, alpha = 0.05)

Arguments

Value

An "easystat_result" object with:

coefficients_table

Observed vs. expected frequency table

model_fit_table

Chi-square statistic, df, p-value, Cramér's V

explanation

Plain-language narrative

Examples

# Independence test
result <- easy_chisq(~ cyl + am, data = mtcars)
print(result)

# Goodness-of-fit
result <- easy_chisq(~ cyl, data = mtcars)
print(result)

Correlation Analysis with Automated Narrative Reporting

Description

Computes bivariate or pairwise correlations (Pearson, Spearman, or Kendall) with significance tests and confidence intervals. For two variables a full narrative is generated; for multiple variables a correlation matrix is returned with a summary digest.

Usage

easy_correlation(
  x,
  y = NULL,
  data = NULL,
  vars = NULL,
  method = "pearson",
  conf_level = 0.95,
  alpha = 0.05
)

Arguments

Value

An "easystat_result" object.

Examples

result <- easy_correlation(~ mpg + wt, data = mtcars)
print(result)

result <- easy_correlation(mtcars, vars = c("mpg", "hp", "wt", "disp"))
print(result)

Correlation Matrix Heatmap

Description

Computes pairwise correlations and displays them as a color-coded heatmap, annotating each cell with the correlation coefficient and a significance star.

Usage

easy_correlation_heatmap(data, vars = NULL, method = "pearson", title = NULL)

Arguments

Value

An "easystat_result" object with plot_object.

Kernel Density Plot with Optional Group Overlay

Description

Draws a smooth kernel density estimate for a numeric variable. If a grouping variable is provided, separate overlapping density curves are drawn per group.

Usage

easy_density(x, data = NULL, group_by = NULL, fill_alpha = 0.35, title = NULL)

Arguments

Value

An "easystat_result" object with plot_object.

Comprehensive Descriptive Statistics with Narrative

Description

Computes a rich set of descriptive statistics for one or more numeric variables, including measures of central tendency, dispersion, shape (skewness, kurtosis), and normality (Shapiro-Wilk), together with an automatic plain-language narrative interpretation.

Usage

easy_describe(data, vars = NULL, digits = 4, conf_level = 0.95)

Arguments

Value

An "easystat_result" object with:

coefficients_table

Wide-format summary statistics table

model_fit_table

Shape and normality digest

explanation

Plain-language narrative (one per variable)

Examples

result <- easy_describe(mtcars, vars = c("mpg", "hp", "wt"))
print(result)

F-Test for Equality of Variances with Automated Narrative Reporting

Description

Performs an F-test to compare the variances of two independent groups using stats::var.test(), extracts the F-statistic, degrees of freedom, p-value, variance ratio, and confidence interval, and generates a plain-language narrative that includes a practical recommendation for downstream t-test selection (equal vs. unequal variances).

Usage

easy_ftest(
  x,
  y = NULL,
  data = NULL,
  ratio = 1,
  alternative = "two.sided",
  conf_level = 0.95,
  alpha = 0.05
)

Arguments

Value

An "easystat_result" object.

Examples

result <- easy_ftest(mpg ~ am, data = mtcars)
print(result)

Group-Wise Summary Statistics with Narrative

Description

Computes descriptive statistics for a numeric outcome variable stratified by a grouping factor, providing both a comparison table and a narrative highlighting which group has the highest/lowest mean and variability.

Usage

easy_group_summary(formula, data, digits = 4)

Arguments

Value

An "easystat_result" object.

Examples

result <- easy_group_summary(mpg ~ cyl, data = mtcars)
print(result)

Annotated Histogram with Normal Curve Overlay

Description

Draws a histogram of a numeric variable, overlays a fitted normal density curve, and annotates the plot with the mean, median, and standard deviation. Normality is assessed via the Shapiro-Wilk test, and the result is displayed in the subtitle.

Usage

easy_histogram(
  x,
  data = NULL,
  bins = NULL,
  fill_color = NULL,
  show_normal = TRUE,
  title = NULL
)

Arguments

Value

An "easystat_result" object with plot_object.

Kruskal-Wallis Test with Automated Narrative Reporting

Description

Runs a Kruskal-Wallis rank-sum test for comparing three or more groups.

Usage

easy_kruskal(formula, data, alpha = 0.05)

Arguments

Value

An "easystat_result" object.

Examples

result <- easy_kruskal(Sepal.Length ~ Species, data = iris)
print(result)

Run a Logistic Regression with Automated Narrative Reporting

Description

Executes a binary logistic regression using stats::glm() with family = binomial, extracts coefficients, odds ratios, approximate confidence intervals, and model-fit metrics, then generates a plain-language narrative via the Narrative Generator Module.

Usage

easy_logistic_regression(formula, data, alpha = 0.05, conf_level = 0.95)

Arguments

Value

An object of class "easystat_result" with coefficient, odds-ratio, model-fit, raw glm, and narrative components.

Examples

result <- easy_logistic_regression(am ~ mpg + wt, data = mtcars)
print(result)

Odds Ratio Plot for Logistic Regression

Description

Creates a coefficient figure showing odds ratios and confidence intervals for easy_logistic_regression results.

Usage

easy_odds_ratio_plot(result)

Arguments

Value

An "easystat_result" object with plot_object.

Normal Q-Q Plot with Shapiro-Wilk Annotation

Description

Plots sample quantiles against theoretical normal quantiles and annotates the Shapiro-Wilk p-value. Deviations from the diagonal indicate non-normality.

Usage

easy_qqplot(x, data = NULL, title = NULL)

Arguments

Value

An "easystat_result" object with plot_object.

Run a Linear Regression with Automated Narrative Reporting

Description

Executes a standard OLS linear regression using stats::lm(), extracts key metrics via the broom package, and automatically generates a plain-language narrative explanation via the Narrative Generator Module.

Usage

easy_regression(formula, data, alpha = 0.05)

Arguments

Value

An object of class "easystat_result" (an R list) with:

test_type

Character: "regression"

formula_str

Character string of the formula used

raw_model

The raw lm object for advanced use

coefficients_table

Tidy data frame of coefficients, SEs, t-stats, p-values

model_fit_table

Data frame with R^2, Adjusted R^2, F-statistic, p-value

explanation

Plain-language narrative string

Examples

result <- easy_regression(mpg ~ wt + hp, data = mtcars)
print(result)

Regression Diagnostic Plot

Description

Creates a fitted-vs-residuals diagnostic figure for linear regression results returned by easy_regression.

Usage

easy_regression_diagnostics(result)

Arguments

Value

An "easystat_result" object with plot_object.

Scatter Plot with Regression Line and Correlation Annotation

Description

Draws a scatter plot for two numeric variables, overlays a linear regression line with confidence band, and annotates the Pearson r and p-value.

Usage

easy_scatter(
  formula,
  data,
  color_by = NULL,
  smooth = TRUE,
  ellipse = TRUE,
  title = NULL
)

Arguments

Value

An "easystat_result" object with plot_object.

Run an Independent-Samples t-Test with Automated Narrative Reporting

Description

Executes a two-sample (or one-sample) t-test using stats::t.test(), extracts key metrics via broom, and generates a plain-language narrative via the Narrative Generator Module.

Usage

easy_ttest(
  x,
  y = NULL,
  data = NULL,
  mu = 0,
  var.equal = FALSE,
  conf.level = 0.95,
  alpha = 0.05
)

Arguments

Value

An object of class "easystat_result" with:

test_type

Character: "ttest"

formula_str

Description of the comparison

raw_model

The raw htest object

coefficients_table

Group means and confidence interval

model_fit_table

t-statistic, df, and p-value

explanation

Plain-language narrative string

Examples

result <- easy_ttest(mpg ~ am, data = mtcars)
print(result)

Wilcoxon Tests with Automated Narrative Reporting

Description

Runs a one-sample, paired, or two-sample Wilcoxon test using stats::wilcox.test() and returns an "easystat_result" object.

Usage

easy_wilcox(
  x,
  y = NULL,
  data = NULL,
  mu = 0,
  paired = FALSE,
  alternative = "two.sided",
  conf_level = 0.95,
  alpha = 0.05
)

Arguments

Value

An "easystat_result" object.

Examples

result <- easy_wilcox(mpg ~ am, data = mtcars)
print(result)

One-Sample and Two-Sample Z-Tests with Automated Narrative Reporting

Description

Performs a z-test using the normal distribution. When the population standard deviation (sigma) is not provided, the sample SD is used (valid for large samples, n \geq 30, by the Central Limit Theorem). Key metrics — z-statistic, p-value, confidence interval, and Cohen's d — are extracted and fed to the Narrative Generator Module.

Usage

easy_ztest(
  x,
  y = NULL,
  data = NULL,
  mu = 0,
  sigma = NULL,
  sigma2 = NULL,
  alternative = "two.sided",
  conf_level = 0.95,
  alpha = 0.05
)

Arguments

Value

An "easystat_result" object.

Examples

# One-sample z-test (large n, CLT)
result <- easy_ztest(mtcars$mpg, mu = 20)
print(result)

# Two-sample z-test via formula
result <- easy_ztest(mpg ~ am, data = mtcars)
print(result)

Export an EasyStat Result to a Formatted Microsoft Word Document

Description

Takes a unified easystat_result object and writes a fully formatted .docx report using the flextable and officer packages. The report contains a title page header, the plain-language narrative, both statistical tables rendered as professional flextable objects, and a footer with metadata. All of this is produced in a single function call.

Usage

export_to_word(result, file, title = NULL, author = "EasyStat")

Arguments

Value

The file path invisibly (a length-one character vector). The .docx file is written to the path supplied via file as a side-effect.

Examples

  result <- easy_regression(mpg ~ wt + hp, data = mtcars)
  export_to_word(result, file = tempfile(fileext = ".docx"),
                author = "Mr. Mahesh Divakaran")

Print an EasyStat Result Object

Description

Automatically renders an easystat_result object. In an interactive RStudio session the HTML Viewer is used; otherwise clean ASCII tables and the plain-language narrative are written to the console.

Usage

## S3 method for class 'easystat_result'
print(x, viewer = NULL, ...)

Arguments

Value

x invisibly.

Summarize an EasyStat Result Object

Description

Summarize an EasyStat Result Object

Usage

## S3 method for class 'easystat_result'
summary(object, ...)

Arguments

Value

Called for its side effects (printing to the console or RStudio Viewer). Returns object invisibly via print.easystat_result.

Apply EasyStat ggplot2 Theme

Description

Adds a clean, professional EasyStat visual theme to any ggplot2 object.

Usage

theme_easystat(base_size = 12, legend_position = "right")

Arguments

Value

A ggplot2::theme object.