### Install statsExpressions Package (Development)

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

Use this command to install the development version of the statsExpressions package using the pak package manager.

```r
pak::pak("IndrajeetPatil/statsExpressions")
```

--------------------------------

### Install statsExpressions Package (Release)

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

Use this command to install the stable release version of the statsExpressions package from CRAN.

```r
install.packages("statsExpressions")
```

--------------------------------

### Check System Dependencies for statsExpressions

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

Run this command on Linux to check for any additional system dependencies required for installing the statsExpressions package.

```r
pak::pkg_sysreqs("statsExpressions")
```

--------------------------------

### Perform Grouped One-Sample T-test with dplyr

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

This example shows how to perform a one-sample t-test for all levels of a grouping variable using dplyr for grouped operations. Ensure dplyr is loaded.

```r
# for reproducibility
set.seed(123)
library(dplyr)

# grouped operation

```

--------------------------------

### Summarize Multiple Data Frames with purrr::map

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/long-to-wide-converter.md

Apply the summary function to a list of data frames using purrr::map to get statistical summaries for each. This is useful for understanding data distributions.

```r
purrr::map(list(df1, df2, df3, df4), summary)
```

--------------------------------

### Get Column Names of Between-Subjects DataFrame

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/oneway-anova-bayes.md

Use `colnames()` to retrieve the names of all columns in a data frame prepared for Bayesian ANOVA.

```R
colnames(df_between)
```

--------------------------------

### Perform Kruskal-Wallis Test and Create Ridgeplot in R

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/paper/JOSS files/paper.md

This snippet shows how to perform a non-parametric one-way ANOVA (Kruskal-Wallis test) using the `oneway_anova` function from the `statsExpressions` package and then visualize the results with a ridgeplot using `ggplot2` and `ggridges`. Ensure the necessary libraries are installed and loaded.

```r
# needed libraries
library(statsExpressions)
library(ggplot2)
library(ggridges)
library(palmerpenguins) # `penguins` dataset is from this package

# creating a dataframe
res <- oneway_anova(penguins, species, body_mass_g, type = "nonparametric")

# create a ridgeplot using `ggridges` package
ggplot(penguins, aes(x = body_mass_g, y = species)) +
  geom_density_ridges(
    jittered_points = TRUE,
    quantile_lines = TRUE,
    scale = 0.9,
    vline_size = 1,
    vline_color = "red",
    position = position_raincloud(adjust_vlines = TRUE)
  ) + # use 'expression' column to display results in the subtitle
  labs(
    x = "Penguin species",
    y = "Body mass (in grams)",
    title = "Kruskal-Wallis Rank Sum Test",
    subtitle = res$expression[[1]]
  )
```

--------------------------------

### Get Column Names of Within-Subjects DataFrame

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/oneway-anova-bayes.md

Use `colnames()` to retrieve the names of all columns in a data frame prepared for within-subjects Bayesian ANOVA.

```R
colnames(df_within)
```

--------------------------------

### Run meta-analysis

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/meta-random-parametric.md

Executes meta-analysis on a dataset, noting that missing required arguments will trigger an error.

```R
meta_analysis(mtcars)
```

--------------------------------

### Select Columns for Meta-Analysis Summary

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/meta-random-robust.md

Use `select` to keep relevant columns for a meta-analysis summary. This is useful for preparing data for further analysis or display.

```R
select(df, -expression)
```

--------------------------------

### Get Number of Rows in Between-Subjects DataFrame

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/oneway-anova-bayes.md

Use `nrow()` to find the total number of rows in a data frame used for Bayesian ANOVA.

```R
nrow(df_between)
```

--------------------------------

### Citation for statsExpressions Package

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

This command displays the citation information for the statsExpressions package, including a BibTeX entry for LaTeX users.

```r
citation("statsExpressions")
```

--------------------------------

### Get Number of Rows in Within-Subjects DataFrame

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/oneway-anova-bayes.md

Use `nrow()` to find the total number of rows in a data frame used for within-subjects Bayesian ANOVA.

```R
nrow(df_within)
```

--------------------------------

### One-Sample Test (Bayes, g, 95% CI) - Inspecting Results

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/one-sample.md

Shows how to inspect the names of the results object from a Bayesian one-sample test. This is useful for understanding the available output fields before accessing specific values.

```r
names(res)
```

--------------------------------

### One-Sample Test (Bayes, g, 95% CI) - Accessing BF10

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/one-sample.md

Demonstrates accessing the Bayes Factor (BF10) from the results of a Bayesian one-sample test. BF10 quantifies the evidence for the alternative hypothesis over the null hypothesis.

```r
res$bf10[[1L]]
```

--------------------------------

### Pairwise Comparisons Output (Yuen's Trimmed Means)

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/pairwise-comparisons.md

Presents the results of pairwise comparisons using Yuen's trimmed means with Hommel adjustment. The output includes group comparisons, estimates, confidence intervals, p-values, and adjusted p-values.

```r
df3
```

--------------------------------

### Load ggplot2 Library for Visualization

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

Loads the ggplot2 library, which is necessary for creating plots with the StatsExpressions package.

```r
library(ggplot2)
```

--------------------------------

### Perform One-Sample Robust T-test and Format Output

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

This snippet demonstrates performing a one-sample robust t-test and then selecting specific columns before formatting the output using knitr::kable. The 'expression' column is excluded as it's not needed for dataframe usage.

```r
set.seed(123)

# one-sample robust t-test
# we will leave `expression` column out; it's not needed for using only the dataframe
mtcars |> 
  one_sample_test(wt, test.value = 3, type = "robust") |> 
  dplyr::select(-expression) |>
  knitr::kable()
```

--------------------------------

### Perform One-Way ANOVA with statsExpressions

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/paper/JOSS files/paper.md

Demonstrates running non-parametric and robust one-way ANOVA tests by toggling the type argument.

```r
mtcars %>% oneway_anova(cyl, wt, type = "nonparametric")
#> # A tibble: 1 x 14
#>   parameter1 parameter2 statistic df.error   p.value
#>   <chr>      <chr>          <dbl>    <int>     <dbl>
#> 1 wt         cyl             22.8        2 0.0000112
#>   method                       estimate conf.level conf.low conf.high
#>   <chr>                           <dbl>      <dbl>    <dbl>     <dbl>
#> 1 Kruskal-Wallis rank sum test    0.736       0.95    0.613     0.831
#>   effectsize      conf.method conf.iterations expression
#>   <chr>           <chr>                 <int> <list>
#> 1 Epsilon2 (rank) bootstrap               100 <language>

mtcars %>% oneway_anova(cyl, wt, type = "robust")
#> # A tibble: 1 x 11
#>   statistic    df df.error p.value estimate conf.level conf.low conf.high
#>       <dbl> <dbl>    <dbl>   <dbl>    <dbl>      <dbl>    <dbl>     <dbl>
#> 1      12.7     2     12.2 0.00102     1.05       0.95    0.843      1.50
#>   effectsize
#>   <chr>
#> 1 Explanatory measure of effect size
#>   method                                            expression
#>   <chr>                                             <list>
#> 1 A heteroscedastic one-way ANOVA for trimmed means <language>
```

--------------------------------

### Basic Pairwise Contingency Table

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/pairwise-contingency-table.md

Generates a basic pairwise contingency table. Use this for a general overview of pairwise comparisons without specific adjustment methods applied to p-values.

```r
select(df1, -expression)
```

```r
df1[["expression"]]
```

--------------------------------

### Pairwise Comparisons Output (Bayesian t-test)

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/pairwise-comparisons.md

Illustrates the output for pairwise comparisons using a Bayesian t-test. This includes group comparisons, effect sizes, estimates, confidence intervals, and terms.

```r
df4
```

--------------------------------

### Contingency Table Analysis with Pie Chart and Caption

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

Analyzes categorical data from 'mpg$class' using a one-sample Bayesian test and visualizes the distribution with a pie chart. The caption displays the parsed statistical expression. Requires 'mpg' data and 'results_data$expression'.

```r
set.seed(123)

# dataframe with results
results_data <- contingency_table(
  as.data.frame(table(mpg$class)),
  Var1,
  counts = Freq,
  type = "bayes"
)

# create a pie chart
ggplot(as.data.frame(table(mpg$class)), aes(x = "", y = Freq, fill = factor(Var1))) +
  geom_bar(width = 1, stat = "identity") +
  theme(axis.line = element_blank()) +
  # cleaning up the chart and adding results from one-sample proportion test
  coord_polar(theta = "y", start = 0) +
  labs(
    fill = "Class",
    x = NULL,
    y = NULL,
    title = "Pie Chart of class (type of car)",
    caption = parse(text = results_data$expression)
  )
```

--------------------------------

### Yuen's Trimmed Means Pairwise Comparisons

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/pairwise-comparisons.md

Shows pairwise comparisons for Yuen's trimmed means test. The output includes estimates, confidence intervals, p-values, and the formatted expression.

```R
df3
```

```R
df3[["expression"]]
```

--------------------------------

### Inspect Data Frames with purrr::walk

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/long-to-wide-converter.md

Use purrr::walk to apply dplyr::glimpse to a list of data frames for inspecting their structure. Ensure purrr and dplyr are loaded.

```r
purrr::walk(list(df1, df2, df3, df4), dplyr::glimpse)
```

--------------------------------

### Perform Meta-Analysis

Source: https://context7.com/indrajeetpatil/statsexpressions/llms.txt

Combines effect sizes from multiple studies using random-effects models. Requires estimate and standard error columns.

```r
library(statsExpressions)
set.seed(123)

# Prepare data: requires 'estimate' and 'std.error' columns
data(mag, package = "metaplus")
dat <- dplyr::rename(mag, estimate = yi, std.error = sei)

# Parametric random-effects meta-analysis (requires metafor package)
meta_analysis(dat, type = "parametric")
#> # A tibble: 1 × 11
#>   term   estimate std.error conf.level conf.low conf.high statistic p.value
#>   <chr>     <dbl>     <dbl>      <dbl>    <dbl>     <dbl>     <dbl>   <dbl>
#> 1 Overall  -0.096     0.044       0.95   -0.182    -0.009     -2.18   0.029
#>   method                                     effectsize             n.obs
#>   <chr>                                      <chr>                  <int>
#> 1 Random-effects meta-analysis (REML method) meta-analytic...          6

# Robust meta-analysis (requires metaplus package)
meta_analysis(dat, type = "robust", random = "normal")

# Bayesian meta-analysis (requires metaBMA package)
meta_analysis(dat, type = "bayes")

# Adjust confidence level
meta_analysis(dat, conf.level = 0.99)
```

--------------------------------

### One-Sample Test with Histogram and Subtitle

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

Performs a one-sample test on 'mtcars' dataset and visualizes the results with a histogram, including a vertical line at the mean and a subtitle parsed from the test results. Requires 'mtcars' data and 'results_data$expression'.

```r
set.seed(123)

# dataframe with results
results_data <- one_sample_test(mtcars, wt, test.value = 3, type = "bayes")

# creating a histogram plot
ggplot(mtcars, aes(wt)) +
  geom_histogram(alpha = 0.5) +
  geom_vline(xintercept = mean(mtcars$wt), color = "red") +
  labs(subtitle = parse(text = results_data$expression))
```

--------------------------------

### Select Columns from DataFrame (With Missing Data)

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/centrality-description.md

This snippet shows how to select columns from a dataframe containing missing values (NA), excluding the 'expression' column. This is useful for data preprocessing when dealing with incomplete datasets.

```r
select(df_na, -expression)
```

--------------------------------

### Displaying pairwise comparison tibbles

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/pairwise-comparisons.md

Displays the resulting tibbles from various statistical tests performed on between-subjects designs.

```R
df1
```

```R
df2
```

```R
df3
```

```R
df4
```

--------------------------------

### Bayesian t-test Pairwise Comparisons

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/pairwise-comparisons.md

Demonstrates pairwise comparisons for a Bayesian t-test. This snippet includes effect sizes, Bayes factors, and formatted expressions for reporting.

```R
df4
```

```R
df4[["expression"]]
```

--------------------------------

### Within-Subjects Two-Sample Test (Non-Parametric) with Tidy Data

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

Prepares data for a within-subjects two-sample test by pivoting it into a long format, then performs a non-parametric test using `two_sample_test` with `paired = TRUE` and `type = "np"`.

```r
set.seed(123)
library(tidyr)
library(PairedData)
data(PrisonStress)

# get data in tidy format
df <- pivot_longer(PrisonStress, starts_with("PSS"), names_to = "PSS", values_to = "stress")

results_data <- two_sample_test(
  data = df,
  x = PSS,
  y = stress,
  paired = TRUE,
  subject.id = Subject,
  type = "np"
)
```

--------------------------------

### Formatted Expressions for p-values (Yuen's Trimmed Means)

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/pairwise-comparisons.md

Retrieves and shows the formatted statistical expressions for the adjusted p-values from the Yuen's trimmed means test. These expressions are ready for use in statistical reporting.

```r
df3[["expression"]]
```

--------------------------------

### Display Mean for Each Level of 'cyl' with Parsed Expressions

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

Generates a plot showing the mean of 'wt' for each 'cyl' level, using `centrality_description` and parsing the 'expression' column for display.

```r
# displaying mean for each level of `cyl`
centrality_description(mtcars, cyl, wt) |>
  ggplot(aes(cyl, wt)) +
  geom_point() +
  geom_label(aes(label = expression), parse = TRUE)
```

--------------------------------

### Student's t-test Pairwise Comparisons

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/pairwise-comparisons.md

Generates pairwise comparisons for a within-subjects design using Student's t-test. The output includes p-values and formatted expressions for reporting.

```R
df1
```

```R
df1[["expression"]]
```

--------------------------------

### One-Way ANOVA with Robust Type and Ridgeplot Visualization

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

Performs a robust one-way ANOVA on 'Sepal.Length' by 'Species' and visualizes the results using a ridgeplot, displaying the formatted expression in the subtitle.

```r
set.seed(123)
library(ggridges)

results_data <- oneway_anova(iris, Species, Sepal.Length, type = "robust")

# create a ridgeplot
ggplot(iris, aes(x = Sepal.Length, y = Species)) +
  geom_density_ridges() +
  labs(
    title = "A heteroscedastic one-way ANOVA for trimmed means",
    subtitle = results_data$expression[[1]]
  )
```

--------------------------------

### Generate Welch ANOVA subtitles

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/oneway-anova-parametric.md

Extracts statistical results and expression objects for Welch's ANOVA.

```R
select(df, -expression)
```

```R
df[["expression"]]
```

--------------------------------

### Perform Pairwise Comparisons

Source: https://context7.com/indrajeetpatil/statsexpressions/llms.txt

Executes post-hoc tests between group pairs. Supports various test types and multiple testing corrections.

```r
library(statsExpressions)
set.seed(123)

# Between-subjects parametric: Games-Howell test (unequal variances)
pairwise_comparisons(mtcars, cyl, wt, type = "parametric")
#> # A tibble: 3 × 8
#>   group1 group2 statistic p.value p.adjust.method test         expression
#>   <chr>  <chr>      <dbl>   <dbl> <chr>           <chr>        <list>
#> 1 4      6          -3.06  0.0282 Holm            Games-Howell <language>
#> 2 4      8          -7.37  0.0003 Holm            Games-Howell <language>
#> 3 6      8          -3.60  0.0195 Holm            Games-Howell <language>

# Student's t-test (equal variances assumed)
pairwise_comparisons(mtcars, cyl, wt, type = "parametric", var.equal = TRUE)

# Non-parametric Dunn test
pairwise_comparisons(mtcars, cyl, wt, type = "nonparametric")

# Robust Yuen's trimmed means test
pairwise_comparisons(mtcars, cyl, wt, type = "robust")

# Bayesian pairwise comparisons
pairwise_comparisons(mtcars, cyl, wt, type = "bayes")

# Custom p-value adjustment method
pairwise_comparisons(mtcars, cyl, wt, p.adjust.method = "bonferroni")
pairwise_comparisons(mtcars, cyl, wt, p.adjust.method = "fdr")
pairwise_comparisons(mtcars, cyl, wt, p.adjust.method = "none")

# Within-subjects (paired) pairwise comparisons
pairwise_comparisons(
  bugs_long, condition, desire,
  subject.id = subject,
  paired = TRUE,
  p.adjust.method = "BH"
)

# Within-subjects non-parametric (Durbin-Conover test)
pairwise_comparisons(
  bugs_long, condition, desire,
  subject.id = subject,
  paired = TRUE,
  type = "nonparametric"
)
```

--------------------------------

### Perform One-Sample Tests with statsExpressions

Source: https://context7.com/indrajeetpatil/statsexpressions/llms.txt

Use `one_sample_test` for one-sample hypothesis tests. Supports parametric, non-parametric, robust, and Bayesian methods. Specify `effsize.type` for custom effect sizes like Cohen's d.

```r
library(statsExpressions)
set.seed(123)

# Parametric one-sample t-test with Hedge's g effect size
one_sample_test(mtcars, wt, test.value = 3)

```

```r
# Non-parametric Wilcoxon test
one_sample_test(mtcars, wt, test.value = 3, type = "nonparametric")
```

```r
# Robust bootstrap-t method for trimmed means
one_sample_test(mtcars, wt, test.value = 3, type = "robust")
```

```r
# Bayesian one-sample t-test with default Cauchy prior (r = 0.707)
one_sample_test(mtcars, wt, test.value = 3, type = "bayes")
```

```r
# Custom effect size (Cohen's d instead of Hedge's g)
one_sample_test(mtcars, wt, test.value = 3, effsize.type = "d")
```

```r
# Adjust confidence level and alternative hypothesis
one_sample_test(mtcars, wt, test.value = 3, conf.level = 0.99, alternative = "greater")
```

--------------------------------

### Contingency Table Expression Invariance

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/contingency-table.md

Demonstrates that the contingency_table expression output is invariant to the `options(OutDec)` setting. This ensures consistent results regardless of decimal separator settings.

```R
df[["expression"]]
```

```R
df[["expression"]]
```

--------------------------------

### Inspect Bayesian One-Sample Test Results

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/one-sample.md

Access the names of the result object or specific Bayes factor values for Bayesian one-sample tests.

```R
names(res)
```

```R
res$bf10[[1L]]
```

--------------------------------

### Meta-Analysis Forest Plot with Custom Subtitle

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

Generates a meta-analysis forest plot using the 'metaviz' package and displays a custom subtitle parsed from the meta-analysis results. Requires 'mozart' data and 'results_data$expression'.

```r
set.seed(123)
library(metaviz)

# dataframe with results
results_data <- meta_analysis(dplyr::rename(mozart, estimate = d, std.error = se))

# meta-analysis forest plot with results random-effects meta-analysis
suppressWarnings(viz_forest(
  x = mozart[, c("d", "se")],
  study_labels = mozart[, "study_name"],
  xlab = "Cohen's d",
  variant = "thick",
  type = "cumulative"
)) + 
  labs(
    title = "Meta-analysis of Pietschnig, Voracek, and Formann (2010) on the Mozart effect",
    subtitle = parse(text = results_data$expression)
  ) +
  theme(text = element_text(size = 12))
```

--------------------------------

### Two-Sample Welch's t-test with Boxplot and Parsed Expression

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

Performs a two-sample Welch's t-test on 'len' by 'supp' using `two_sample_test` and displays the parsed expression as the subtitle of a boxplot.

```r
set.seed(123)

results_data <- two_sample_test(ToothGrowth, supp, len)

ggplot(ToothGrowth, aes(supp, len)) +
  geom_boxplot() +
  labs(
    title = "Two-Sample Welch's t-test",
    subtitle = parse(text = results_data$expression)
  )
```

--------------------------------

### Perform One-Way ANOVA (Non-parametric)

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

Use this function to perform a non-parametric one-way ANOVA. Ensure the data is piped into the function.

```r
mtcars |> oneway_anova(cyl, wt, type = "nonparametric")
```

--------------------------------

### Inspect Parametric One-Sample Test Results

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/one-sample.md

View the tibble of results excluding the expression column or access the expression object directly.

```R
select(res, -expression)
```

```R
res[["expression"]]
```

--------------------------------

### pairwise_comparisons

Source: https://context7.com/indrajeetpatil/statsexpressions/llms.txt

Performs pairwise post-hoc comparisons between all group pairs following a one-way design, supporting various parametric, non-parametric, robust, and Bayesian tests.

```APIDOC
## pairwise_comparisons

### Description
Performs pairwise post-hoc comparisons between all group pairs following a one-way design. Supports parametric (Games-Howell or Student's t), non-parametric (Dunn or Durbin-Conover), robust (Yuen's trimmed means), and Bayesian (BayesFactor) tests. Includes multiple testing correction methods.

### Parameters
- **data** (data.frame) - Required - The dataset containing the variables.
- **x** (variable) - Required - The grouping variable.
- **y** (variable) - Required - The numeric response variable.
- **type** (string) - Optional - Test type: "parametric", "nonparametric", "robust", or "bayes".
- **p.adjust.method** (string) - Optional - Correction method (e.g., "holm", "bonferroni", "fdr").
- **var.equal** (boolean) - Optional - Whether to assume equal variances for parametric tests.
- **subject.id** (variable) - Optional - Subject identifier for within-subjects designs.
- **paired** (boolean) - Optional - Whether the design is paired (within-subjects).

### Response
- Returns a tibble containing group comparisons, statistics, p-values, and adjustment methods.
```

--------------------------------

### Perform One-Way ANOVA (Robust)

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

Use this function to perform a robust one-way ANOVA. Ensure the data is piped into the function.

```r
mtcars |> oneway_anova(cyl, wt, type = "robust")
```

--------------------------------

### centrality_description

Source: https://context7.com/indrajeetpatil/statsexpressions/llms.txt

Computes centrality measures (mean, median, trimmed mean, or MAP) for a numeric variable across groups.

```APIDOC
## centrality_description

### Description
Computes centrality measures for a numeric variable across groups. Returns a data frame with distribution statistics and pre-formatted expressions for labeling plots.

### Parameters
- **data** (data.frame) - Required - The dataset.
- **x** (variable) - Required - The grouping variable.
- **y** (variable) - Required - The numeric variable to summarize.
- **type** (string) - Optional - Centrality type: "parametric" (mean), "nonparametric" (median), "robust" (trimmed mean), or "bayes" (MAP).
- **tr** (numeric) - Optional - Trim level for robust centrality.
- **conf.level** (numeric) - Optional - Confidence level for intervals.

### Response
- Returns a tibble with distribution statistics (iqr, min, max, skewness, kurtosis, n.obs) and formatted expressions.
```

--------------------------------

### Perform Correlation Test

Source: https://context7.com/indrajeetpatil/statsexpressions/llms.txt

Use corr_test to analyze the correlation between two continuous variables. Supports Pearson, Spearman, robust (Winsorized Pearson), and Bayesian correlation coefficients, all returning confidence intervals.

```r
library(statsExpressions)
set.seed(123)

# Parametric Pearson correlation
corr_test(mtcars, wt, mpg, type = "parametric")
```

```r
# Non-parametric Spearman correlation
corr_test(mtcars, wt, mpg, type = "nonparametric")
```

```r
# Robust Winsorized correlation
corr_test(mtcars, wt, mpg, type = "robust")
```

```r
# Bayesian correlation with HDI intervals
corr_test(mtcars, wt, mpg, type = "bayes")
```

```r
# Adjust confidence level
corr_test(mtcars, wt, mpg, conf.level = 0.99)
```

```r
# Adjust trim level for robust correlation
corr_test(mtcars, wt, mpg, type = "robust", tr = 0.1)
```

--------------------------------

### Extracting Contingency Table Expression without Plotting

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

Demonstrates how to extract the statistical expression for a contingency table test (Pearson's chi-squared) without generating a plot. This is useful for using the expression in other contexts. Requires 'mtcars' data.

```r
set.seed(123)

# Pearson's chi-squared test of independence
contingency_table(mtcars, am, vs)$expression[[1]]
```

--------------------------------

### Select meta-analysis results

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/meta-random-bayes.md

Removes the expression column from the data frame to display the primary meta-analytic results.

```R
dplyr::select(df, -expression)
```

--------------------------------

### One Sample t-test Output

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/one-two-sample-dataframes.md

Displays the tibble output for a one-sample t-test analysis.

```R
df_1
```

--------------------------------

### meta_analysis

Source: https://context7.com/indrajeetpatil/statsexpressions/llms.txt

Performs random-effects meta-analysis combining effect sizes from multiple studies.

```APIDOC
## meta_analysis

### Description
Performs random-effects meta-analysis combining effect sizes from multiple studies. Requires a data frame with 'estimate' and 'std.error' columns.

### Parameters
- **data** (data.frame) - Required - Data frame containing 'estimate' and 'std.error'.
- **type** (string) - Optional - Approach: "parametric", "robust", or "bayes".
- **random** (string) - Optional - Random effects distribution type (e.g., "normal").
- **conf.level** (numeric) - Optional - Confidence level for intervals.

### Response
- Returns a tibble with meta-analytic results including effect size, confidence intervals, statistics, and p-values.
```

--------------------------------

### Generate Fisher ANOVA subtitles

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/oneway-anova-parametric.md

Extracts statistical results and expression objects for Fisher's ANOVA.

```R
select(df1, -expression)
```

```R
df1[["expression"]]
```

--------------------------------

### Pairwise Comparisons Output (Durbin-Conover)

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/pairwise-comparisons.md

Shows the results of pairwise comparisons using the Durbin-Conover test with BY adjustment. This output includes group comparisons, statistics, p-values, and adjusted p-values.

```r
df2
```

--------------------------------

### Display Data Without NAs Tibble

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/pairwise-comparisons.md

Display the tibble containing results from data without NAs.

```R
df
```

--------------------------------

### Select Data and Expressions for Between-Subjects Analysis

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/oneway-anova-nonparametric.md

Use select to remove the expression column from the tibble or access the expression list directly for between-subjects data.

```R
select(df1, -expression)
```

```R
df1[["expression"]]
```

```R
select(df2, -expression)
```

```R
df2[["expression"]]
```

--------------------------------

### Compute Centrality Measures

Source: https://context7.com/indrajeetpatil/statsexpressions/llms.txt

Calculates group-level summary statistics like mean or median, returning formatted expressions for plot annotations.

```r
library(statsExpressions)
set.seed(123)

# Parametric: Mean for each group
centrality_description(iris, Species, Sepal.Length, type = "parametric")
#> # A tibble: 3 × 13
#>   Species    Sepal.Length   iqr   min   max skewness kurtosis n.obs
#>   <fct>             <dbl> <dbl> <dbl> <dbl>    <dbl>    <dbl> <int>
#> 1 setosa             5.01 0.400   4.3   5.8    0.120   -0.253    50
#> 2 versicolor         5.94 0.700   4.9   7.0    0.105   -0.533    50
#> 3 virginica          6.59 0.675   4.9   7.9    0.118    0.033    50
#>   expression   n.expression std.dev conf.low conf.high
#>   <list>       <chr>          <dbl>    <dbl>     <dbl>
#> 1 <language>   setosa...       0.352     4.91      5.11
#> 2 <language>   versic...       0.516     5.79      6.08
#> 3 <language>   virgin...       0.636     6.41      6.77

# Non-parametric: Median for each group
centrality_description(mtcars, am, wt, type = "nonparametric")

# Robust: Trimmed mean (20% trim by default)
centrality_description(ToothGrowth, supp, len, type = "robust")

# Bayesian: Maximum A Posteriori (MAP) estimate
centrality_description(sleep, group, extra, type = "bayes")

# Adjust trim level for robust centrality
centrality_description(ToothGrowth, supp, len, type = "robust", tr = 0.1)

# Adjust confidence level
centrality_description(iris, Species, Sepal.Length, conf.level = 0.99)
```

--------------------------------

### Perform Two-Sample Tests with statsExpressions

Source: https://context7.com/indrajeetpatil/statsexpressions/llms.txt

Utilize `two_sample_test` for comparing two groups. Handles parametric, non-parametric, robust, and Bayesian approaches for both between-subjects and within-subjects designs. Specify `var.equal = TRUE` for Student's t-test.

```r
library(statsExpressions)
set.seed(123)

# Between-subjects: Welch's t-test (unequal variances)
two_sample_test(ToothGrowth, supp, len, type = "parametric")

```

```r
# Equal variances assumption (Student's t-test)
two_sample_test(ToothGrowth, supp, len, var.equal = TRUE)
```

```r
# Non-parametric Mann-Whitney U test
two_sample_test(ToothGrowth, supp, len, type = "nonparametric")
```

```r
# Robust Yuen's test for trimmed means
two_sample_test(ToothGrowth, supp, len, type = "robust")
```

```r
# Bayesian t-test
two_sample_test(ToothGrowth, supp, len, type = "bayes")
```

```r
# Within-subjects (paired) design
df <- dplyr::filter(bugs_long, condition %in% c("LDLF", "LDHF"))
two_sample_test(df, condition, desire, subject.id = subject, paired = TRUE)
```

```r
# Paired non-parametric (Wilcoxon signed-rank)
two_sample_test(df, condition, desire, subject.id = subject, paired = TRUE, type = "np")
```

```r
# Paired robust test
two_sample_test(df, condition, desire, subject.id = subject, paired = TRUE, type = "robust")
```

--------------------------------

### Display Games-Howell Test Tibble

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/pairwise-comparisons.md

Display the tibble containing Games-Howell test results.

```R
df6
```

--------------------------------

### Access Meta-Analysis Expression Components

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/meta-random-robust.md

Access the 'expression' column from a data frame to retrieve detailed meta-analysis results. This expression contains statistical components like z-values, p-values, and confidence intervals.

```R
df[["expression"]]
```

--------------------------------

### Two Sample Between-Subjects t-test Output

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/one-two-sample-dataframes.md

Displays the tibble output for a between-subjects two-sample t-test.

```R
df_2_between
```

--------------------------------

### Analyze Contingency Tables

Source: https://context7.com/indrajeetpatil/statsexpressions/llms.txt

Use contingency_table for categorical variable analysis. Supports chi-squared test of independence, McNemar's test, and goodness-of-fit tests with frequentist and Bayesian options. Returns effect sizes like Cramer's V or Cohen's g.

```r
library(statsExpressions)
set.seed(123)

# Two-way contingency table: Chi-squared test of independence
contingency_table(mtcars, am, vs)
```

```r
# Bayesian contingency table analysis
contingency_table(mtcars, am, vs, type = "bayes")
```

```r
# Paired (McNemar's test) for repeated measures categorical data
paired_data <- tibble::tibble(
  response_before = factor(c("no", "yes", "no", "yes"), levels = c("no", "yes")),
  response_after = factor(c("no", "no", "yes", "yes"), levels = c("no", "yes")),
  Freq = c(65L, 25L, 5L, 5L)
)
contingency_table(paired_data, response_before, response_after, paired = TRUE, counts = Freq)
```

```r
# One-way goodness-of-fit test
contingency_table(as.data.frame(HairEyeColor), Eye, counts = Freq)
```

```r
# Goodness-of-fit with custom expected proportions
contingency_table(
  as.data.frame(HairEyeColor),
  Eye,
  counts = Freq,
  ratio = c(0.2, 0.2, 0.3, 0.3)
)
```

```r
# Bayesian one-way test
contingency_table(
  as.data.frame(HairEyeColor),
  Eye,
  counts = Freq,
  type = "bayes"
)
```

--------------------------------

### McNemar's Chi-squared Test Results

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/contingency-table.md

Presents the results of McNemar's Chi-squared test for paired nominal data, showing the statistic, degrees of freedom, p-value, effect size (Cohen's g), and confidence intervals. Suitable for analyzing paired categorical data, such as before-and-after measurements.

```R
select(df1, -expression)
```

```R
df1[["expression"]]
```

```R
select(df2, -expression)
```

```R
df2[["expression"]]
```

--------------------------------

### Accessing 'expression' Column (With Missing Data)

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/centrality-description.md

This code accesses the 'expression' column of a dataframe that contains missing data. The output illustrates how the 'expression' column stores statistical results, similar to the non-missing data case, indicating the package's ability to handle NA values.

```r
df_na[["expression"]]
```

--------------------------------

### Access Games-Howell Test Expressions

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/pairwise-comparisons.md

Access the expression column from the dataframe containing Games-Howell test results.

```R
df6[["expression"]]
```

--------------------------------

### Select and Remove Expression from Results

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/centrality-description.md

Use `select(res, -expression)` to display centrality results while excluding the expression column. This is useful for cleaner output when the expression details are not needed.

```R
select(res, -expression)
```

--------------------------------

### Correlation Analysis Scatter Plot with Subtitle

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

Conducts a non-parametric correlation test on 'mtcars' and visualizes the relationship using a scatter plot with a linear model fit. The subtitle is dynamically generated from the test results. Ensure 'mtcars' data and 'results_data$expression' are available.

```r
set.seed(123)

# dataframe with results
results_data <- corr_test(mtcars, mpg, wt, type = "nonparametric")

# create a scatter plot
ggplot(mtcars, aes(mpg, wt)) +
  geom_point() +
  geom_smooth(method = "lm", formula = y ~ x) +
  labs(
    title = "Spearman's rank correlation coefficient",
    subtitle = parse(text = results_data$expression)
  )
```

--------------------------------

### Access Yuen's Trimmed Means Expressions

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/pairwise-comparisons.md

Access the expression column from the dataframe containing Yuen's trimmed means results.

```R
df4[["expression"]]
```

--------------------------------

### One-Sample Test for Grouping Variable Levels

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

Applies a one-sample test to each level of a grouping variable. Requires grouping the data and then applying the test using `group_modify`.

```r
mtcars |>
  group_by(cyl) |>
  group_modify(~ one_sample_test(.x, wt, test.value = 3), .keep = TRUE) |>
  ungroup()
```

--------------------------------

### Paired Test Plot with Custom Subtitle

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

Generates a paired profiles plot for a two-sample Wilcoxon paired test and customizes the subtitle using parsed results. Ensure 'PrisonStress' data and 'results_data$expression' are available.

```r
suppressWarnings(paired.plotProfiles(
  PrisonStress, "PSSbefore", "PSSafter",
  subjects = "Subject"
)) +
  labs(
    title = "Two-sample Wilcoxon paired test",
    subtitle = parse(text = results_data$expression)
  )
```

--------------------------------

### Within-Subjects One-Way ANOVA (Non-Parametric) with Parsed Expressions

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

Conducts a non-parametric within-subjects one-way ANOVA on 'Taste' by 'Wine' for each 'Taster', using `oneway_anova` with `paired = TRUE` and `type = "np"`. The parsed expression is used as the plot subtitle.

```r
set.seed(123)
library(WRS2)
library(ggbeeswarm)

results_data <- oneway_anova(
  WineTasting,
  Wine,
  Taste,
  paired = TRUE,
  subject.id = Taster,
  type = "np"
)

ggplot2::ggplot(WineTasting, aes(Wine, Taste, color = Wine)) +
  geom_quasirandom() +
  labs(
    title = "Friedman's rank sum test",
    subtitle = parse(text = results_data$expression)
  )
```

--------------------------------

### Perform One-Way ANOVA

Source: https://context7.com/indrajeetpatil/statsexpressions/llms.txt

Use oneway_anova for comparing three or more groups. Supports various ANOVA types (Welch's, Fisher's, Kruskal-Wallis, Friedman, robust, Bayesian) and designs (between-subjects, within-subjects). Returns effect sizes like partial omega-squared or Kendall's W.

```r
library(statsExpressions)
set.seed(123)

# Between-subjects: Welch's one-way ANOVA (default, handles unequal variances)
oneway_anova(mtcars, cyl, wt)
```

```r
# Biased effect size (partial eta-squared)
oneway_anova(mtcars, cyl, wt, effsize.type = "eta")
```

```r
# Fisher's ANOVA (equal variances)
oneway_anova(mtcars, cyl, wt, var.equal = TRUE)
```

```r
# Non-parametric Kruskal-Wallis test
oneway_anova(mtcars, cyl, wt, type = "np")
```

```r
# Robust heteroscedastic ANOVA for trimmed means
oneway_anova(mtcars, cyl, wt, type = "robust")
```

```r
# Bayesian ANOVA
oneway_anova(mtcars, cyl, wt, type = "bayes")
```

```r
# Within-subjects (repeated measures) parametric ANOVA
oneway_anova(iris_long, condition, value, subject.id = id, paired = TRUE)
```

```r
# Within-subjects non-parametric (Friedman test)
oneway_anova(iris_long, condition, value, subject.id = id, paired = TRUE, type = "np")
```

```r
# Within-subjects robust ANOVA
oneway_anova(iris_long, condition, value, subject.id = id, paired = TRUE, type = "robust")
```

--------------------------------

### One-way Analysis of Means Output

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/one-two-sample-dataframes.md

Displays the tibble output for a one-way analysis of means.

```R
df_3_between
```

--------------------------------

### Access expression list in z model

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/tidy-model-expressions.md

Retrieves the expression list column from the z-test model data frame.

```R
df_z[["expression"]]
```

--------------------------------

### Select columns excluding expression (within-subjects, no NAs)

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/two-sample-robust.md

This snippet demonstrates selecting all columns from a dataframe except for the 'expression' column in a within-subjects design without missing values.

```r
select(df1, -expression)
```

--------------------------------

### Access Bayesian ANOVA Expression (Between-Subjects)

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/oneway-anova-bayes.md

Access the 'expression' column for detailed Bayesian ANOVA results, including Bayes Factor, R-squared, and credible intervals.

```R
df2[["expression"]][[1]]
```

--------------------------------

### Access Bayesian ANOVA Expression (Between-Subjects)

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/oneway-anova-bayes.md

Access the 'expression' column for detailed Bayesian ANOVA results, including Bayes Factor, R-squared, and credible intervals.

```R
df1[["expression"]][[1]]
```

--------------------------------

### Repeat Statistical Analysis with dplyr

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/paper/JOSS files/paper.md

Combines dplyr grouping with statsExpressions functions to perform tests across multiple levels of a variable.

```r
# running one-sample proportion test for all levels of `cyl`
mtcars %>%
  group_by(cyl) %>%
  group_modify(~ contingency_table(.x, am), .keep = TRUE) %>%
  ungroup()
#> # A tibble: 3 x 13
#>     cyl statistic    df p.value method
#>   <dbl>     <dbl> <dbl>   <dbl> <chr>
#> 1     4     2.27      1 0.132   Chi-squared test for given probabilities
#> 2     6     0.143     1 0.705   Chi-squared test for given probabilities
#> 3     8     7.14      1 0.00753 Chi-squared test for given probabilities
#>   estimate conf.level conf.low conf.high effectsize        conf.method
#>      <dbl>      <dbl>    <dbl>     <dbl> <chr>             <chr>
#> 1    0.344       0.95    0         0.917 Cramer's V (adj.) ncp
#> 2    0           0.95    0         0     Cramer's V (adj.) ncp
#> 3    0.685       0.95    0.127     1.18  Cramer's V (adj.) ncp
#>   conf.distribution expression
#>   <chr>             <list>
#> 1 chisq             <language>
#> 2 chisq             <language>
#> 3 chisq             <language>
```

--------------------------------

### Display Student's t-test Tibble

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/tests/testthat/_snaps/pairwise-comparisons.md

Display the tibble containing Student's t-test results.

```R
df5
```

--------------------------------

### Customizing Statistical Expressions in Plots

Source: https://github.com/indrajeetpatil/statsexpressions/blob/main/README.md

Shows how to extract and selectively use parts of a statistical expression for plot subtitles, focusing on F-statistic and p-value from a one-way ANOVA. Requires 'iris' data and 'res_expr' containing the full expression.

```r
set.seed(123)

# extracting detailed expression
(res_expr <- oneway_anova(iris, Species, Sepal.Length, var.equal = TRUE)$expression[[1]])

# adapting the details to your liking
ggplot(iris, aes(x = Species, y = Sepal.Length)) +
  geom_boxplot() +
  labs(subtitle = ggplot2::expr(paste(
    NULL, italic("F"), "(", "2", ",", "147", ") = ", "119.26", ", ",
    italic("p"), " = ", "1.67e-31"
  )))
```

--------------------------------

### Generate statistical expressions with add_expression_col

Source: https://context7.com/indrajeetpatil/statsexpressions/llms.txt

Creates plotmath-compatible expressions from a data frame of statistical results for use in ggplot2 labels.

```r
library(statsExpressions)
set.seed(123)

# Create a data frame with statistical results
stats_df <- data.frame(
  statistic  = 5.494,
  df         = 29.234,
  p.value    = 0.00001,
  estimate   = -1.980,
  conf.level = 0.95,
  conf.low   = -2.873,
  conf.high  = -1.088,
  method     = "Student's t-test"
)

# Generate expression for t-statistic with Cohen's d effect size
result <- add_expression_col(
  data           = stats_df,
  statistic.text = list(quote(italic("t"))),
  effsize.text   = list(quote(italic("d"))),
  n              = 32L,
  n.text         = list(quote(italic("n")["obs"])),
  digits         = 3L,
  digits.df      = 3L
)

# The expression can be used in ggplot2 subtitle
result$expression[[1]]
#> list(italic("t")(29.234) == "5.494", italic(p) == "0.00001",
#>      italic("d") == "-1.980", CI["95%"] ~ "[" * "-2.873", "-1.088" * "]",
#>      italic("n")["obs"] == "32")

# Use in a ggplot
library(ggplot2)
ggplot(mtcars, aes(wt, mpg)) +
  geom_point() +
  labs(subtitle = result$expression[[1]])
```