Package 'testflow'

Convert a testflow object to a one-row tibble

Description

Convert a testflow object to a one-row tibble

Usage

as_tibble(x, ...)

Arguments

x	A testflow object.
...	Unused.

Value

A one-row tibble with the workflow name, design, variables, recommended test, null hypothesis, statistic, degrees of freedom when available, p-value, confidence interval when available, effect-size fields, and decision text.

---

Simulate a small cardiovascular teaching dataset

Description

Simulate a small cardiovascular teaching dataset

Usage

make_cardio_data(n = 180, seed = 2026)

Arguments

n	Number of rows.
seed	Random seed.

Value

A tibble with example numeric and categorical variables.

---

Plot a testflow object

Description

Plot a testflow object

Usage

## S3 method for class 'testflow'
plot(x, title = NULL, subtitle = NULL, caption = NULL, ...)

Arguments

x	A testflow object.
title	Optional plot title override. Defaults to the stored title.
subtitle	Optional plot subtitle override. Defaults to the stored subtitle.
caption	Optional plot caption override. Defaults to the stored caption.
...	Unused.

Value

A ggplot object stored in the testflow object, with optional title, subtitle, and caption overrides applied. If the workflow was created with plot = FALSE, returns NULL.

---

Print a testflow object

Description

Print a testflow object.

Usage

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

Arguments

x	A testflow object.
...	Unused.

Details

Console colors are enabled by default in interactive sessions. Use options(testflow.cli_colors = FALSE) to disable colors, or options(testflow.cli_colors = TRUE) to force colors in non-interactive output.

Value

The input testflow object, invisibly. Called for its side effect of printing a formatted workflow summary to the console.

---

Return a ready-to-use testflow report

Description

Return a ready-to-use testflow report

Usage

report(x, ...)

Arguments

x	A testflow object.
...	Unused.

Value

A length-one character vector containing a plain-language summary of the workflow result, including the design, recommended test, p-value, effect size when reported, and null hypothesis when available.

---

Return a ready-to-use testflow report

Description

Return a ready-to-use testflow report

Usage

report_test(x)

Arguments

x	A testflow object.

Value

A length-one character vector containing the same report text as report() for a testflow object.

---

Summarize a testflow object

Description

Summarize a testflow object.

Usage

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

Arguments

object	A testflow object.
...	Unused.

Details

Console colors follow the same testflow.cli_colors option used by print.testflow().

Value

A summary.testflow list containing the workflow metadata, descriptives, assumptions, recommended test, primary and alternative test results, post-hoc results when available, effect size, decision, and report text.

---

Build a compact descriptive summary table

Description

Builds a compact table of descriptive summaries using a formula interface. Numeric variables are summarized as mean (SD); median [Q1, Q3]; n. Categorical variables are summarized as n (percent).

Usage

sumtab(
  formula,
  data,
  p_value = FALSE,
  overall = TRUE,
  digits = 1,
  p_digits = 3,
  alpha = 0.05,
  fisher_threshold = 5,
  na.rm = TRUE
)

Arguments

formula	A one-sided formula such as ~ age + sex or ~ age + sex | treatment.
data	A data frame.
p_value	Logical; add a p-value column when a grouping variable is supplied.
overall	Logical; include an overall summary column.
digits	Number of digits for summary statistics.
p_digits	Number of digits for formatted p-values.
alpha	Significance level used by automatic test selection.
fisher_threshold	Expected-count threshold for Fisher's exact test.
na.rm	Logical; remove missing values before summaries and tests.

Details

When p_value = TRUE and a grouping variable is supplied, sumtab() chooses the p-value test automatically. Numeric variables use Student t-test, Welch t-test, or Wilcoxon rank-sum test for two groups, and one-way ANOVA, Welch ANOVA, or Kruskal-Wallis test for more than two groups. Categorical variables use a chi-square test unless expected counts fall below fisher_threshold, in which case Fisher's exact test is used.

Value

A tibble with one row per numeric variable and one row per categorical level.

Examples

dat <- make_cardio_data(80, seed = 1)
sumtab(~ age + sex | treatment, dat, p_value = TRUE)

---

Test association between two categorical variables

Description

Test association between two categorical variables

Usage

test_categorical(
  formula,
  data,
  y = NULL,
  alpha = 0.05,
  fisher_threshold = 5,
  plot = TRUE,
  na.rm = TRUE
)

Arguments

formula	A formula such as x ~ y, or a data frame when using pipe/data-first style.
data	A data frame, or a first categorical column when using data-first style.
y	Second categorical column. Optional when using formula style.
alpha	Significance level.
fisher_threshold	Expected-count threshold for Fisher's exact test.
plot	Logical; include a ggplot object.
na.rm	Logical; remove missing values.

Value

A testflow object with class testflow_categorical. The object is a list containing the cleaned data, categorical descriptives, assumption checks, recommended association test, primary test result with null hypothesis, alternative chi-square and Fisher results, effect size, optional ggplot, original call, and report text.

References

Pearson, K. (1900). On the criterion that a given system of deviations from the probable in the case of a correlated system of variables is such that it can be reasonably supposed to have arisen from random sampling. Philosophical Magazine, 50(302), 157-175.

Fisher, R. A. (1922). On the interpretation of chi-square from contingency tables, and the calculation of P. Journal of the Royal Statistical Society, 85(1), 87-94.

Cramer, H. (1946). Mathematical Methods of Statistics. Princeton.

---

Test correlation between two numeric variables

Description

Test correlation between two numeric variables

Usage

test_correlation(
  formula,
  data,
  y = NULL,
  method = c("auto", "pearson", "spearman", "kendall"),
  alpha = 0.05,
  plot = TRUE,
  na.rm = TRUE
)

Arguments

formula	A formula such as y ~ x, or a data frame when using pipe/data-first style.
data	A data frame, or a first numeric column when using data-first style.
y	Second numeric column. Optional when using formula style.
method	Correlation method or "auto".
alpha	Significance level.
plot	Logical; include a ggplot object.
na.rm	Logical; remove missing values.

Value

A testflow object with class testflow_correlation. The object is a list containing the cleaned complete-case data, numeric descriptives, assumption checks, recommended correlation method, primary correlation test with null hypothesis, Pearson/Spearman/Kendall results, a correlation table, effect size, optional ggplot, original call, and report text.

References

Pearson, K. (1895). Notes on regression and inheritance in the case of two parents. Proceedings of the Royal Society of London, 58, 240-242.

Spearman, C. (1904). The proof and measurement of association between two things. The American Journal of Psychology, 15(1), 72-101.

Kendall, M. G. (1938). A new measure of rank correlation. Biometrika, 30(1/2), 81-93.

---

Test a correlation matrix

Description

Test a correlation matrix

Usage

test_correlation_matrix(
  data,
  vars,
  method = c("spearman", "pearson", "kendall"),
  alpha = 0.05,
  plot = TRUE,
  na.rm = TRUE
)

Arguments

data	A data frame.
vars	Numeric columns.
method	Correlation method.
alpha	Significance level.
plot	Logical; include a ggplot object.
na.rm	Logical; remove missing values.

Value

A testflow object with class testflow_correlation_matrix. The object is a list containing the cleaned data, numeric descriptives, screening assumptions, selected correlation-matrix method, pairwise correlation matrix, pairwise p-value table, maximum absolute correlation as an effect-size summary, optional heatmap ggplot, original call, and report text.

---

Run a factorial ANOVA workflow

Description

Run a factorial ANOVA workflow

Usage

test_factorial(
  formula,
  data,
  factors = NULL,
  alpha = 0.05,
  type = 2,
  plot = TRUE,
  na.rm = TRUE
)

Arguments

formula	A formula such as outcome ~ factor1 * factor2, or a data frame when using data-first style.
data	A data frame, or the outcome column when using data-first style.
factors	Factor columns selected with tidyselect syntax. Optional when using formula style.
alpha	Significance level.
type	ANOVA type placeholder for future car integration.
plot	Logical; include a ggplot object.
na.rm	Logical; remove missing values.

Value

A testflow object with class testflow_factorial. The object is a list containing the cleaned data, descriptive statistics, residual and variance assumption checks, recommended factorial ANOVA, primary ANOVA term result with null hypothesis, ANOVA table, effect size, optional ggplot, original call, and report text.

References

Fisher, R. A. (1925). Statistical Methods for Research Workers. Oliver and Boyd.

Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences (2nd ed.). Lawrence Erlbaum.

---

Compare a numeric outcome across more than two groups

Description

Compare a numeric outcome across more than two groups

Usage

test_groups(
  formula,
  data,
  group = NULL,
  alpha = 0.05,
  posthoc = TRUE,
  plot = TRUE,
  na.rm = TRUE
)

Arguments

formula	A formula such as outcome ~ group, or a data frame when using pipe/data-first style.
data	A data frame, or an outcome column when using data-first style.
group	Grouping column. Optional when using formula style.
alpha	Significance level.
posthoc	Logical; compute post-hoc comparisons.
plot	Logical; include a ggplot object.
na.rm	Logical; remove missing values.

Value

A testflow object with class testflow_groups. The object is a list containing the cleaned data, grouped descriptive statistics, assumption checks, recommended omnibus test, primary test result with null hypothesis, alternative omnibus results, post-hoc comparisons when requested, effect size, optional ggplot, original call, and report text.

---

Test multinomial goodness of fit

Description

Test multinomial goodness of fit

Usage

test_multinomial(
  data,
  outcome,
  p = NULL,
  alpha = 0.05,
  plot = TRUE,
  na.rm = TRUE
)

Arguments

data	A data frame.
outcome	Categorical outcome column.
p	Expected probabilities, or NULL for equal probabilities.
alpha	Significance level.
plot	Logical; include a ggplot object.
na.rm	Logical; remove missing values.

Value

A testflow object with class testflow_multinomial. The object is a list containing the cleaned data, categorical descriptives, assumption checks, recommended goodness-of-fit test, primary chi-square result with null hypothesis, pairwise binomial checks, effect-size summary, optional ggplot, original call, and report text.

References

Pearson, K. (1900). On the criterion that a given system of deviations from the probable in the case of a correlated system of variables is such that it can be reasonably supposed to have arisen from random sampling. Philosophical Magazine, 50(302), 157-175.

---

Test one numeric sample against a reference value

Description

Test one numeric sample against a reference value

Usage

test_one_sample(
  data,
  outcome,
  mu = 0,
  alternative = c("two.sided", "less", "greater"),
  alpha = 0.05,
  plot = TRUE,
  na.rm = TRUE
)

Arguments

data	A data frame.
outcome	Numeric outcome column.
mu	Reference value.
alternative	Alternative hypothesis.
alpha	Significance level.
plot	Logical; include a ggplot object.
na.rm	Logical; remove missing values.

Value

A testflow object with class testflow_one_sample. The object is a list containing the cleaned data, descriptive statistics, assumption checks, recommended test, primary test result with null hypothesis, alternative test results, effect size, optional ggplot, original call, and report text.

References

Gosset, W. S. (1908). The probable error of a mean. Biometrika, 6(1), 1-25.

Wilcoxon, F. (1945). Individual comparisons by ranking methods. Biometrics Bulletin, 1(6), 80-83.

---

Detect numeric outliers

Description

Detect numeric outliers

Usage

test_outliers(
  formula,
  data,
  group = NULL,
  method = c("iqr", "mahalanobis", "both"),
  plot = TRUE,
  na.rm = TRUE
)

Arguments

formula	Numeric columns to screen for outliers.
data	A data frame.
group	Optional grouping column.
method	Outlier method.
plot	Logical; include a ggplot object.
na.rm	Logical; remove missing values.

Value

A testflow object with class testflow_outliers. The object is a list containing the cleaned data, numeric descriptives, screening assumptions, selected outlier-screening method, flagged IQR and/or Mahalanobis rows, outlier-count summary, optional ggplot, original call, and report text. This is a screening workflow, not a single hypothesis test.

---

Compare paired before and after numeric measurements

Description

Compare paired before and after numeric measurements

Usage

test_paired(
  formula,
  data,
  after = NULL,
  alternative = c("two.sided", "less", "greater"),
  alpha = 0.05,
  plot = TRUE,
  na.rm = TRUE
)

Arguments

formula	A formula such as after ~ before, or a data frame when using data-first style.
data	A data frame, or the before column when using data-first style.
after	After column. Optional when using formula style.
alternative	Alternative hypothesis.
alpha	Significance level.
plot	Logical; include a ggplot object.
na.rm	Logical; remove missing values.

Value

A testflow object with class testflow_paired. The object is a list containing the cleaned paired data, paired difference, descriptive statistics, assumption checks, recommended paired test, primary test result with null hypothesis, alternative test results, effect size, optional ggplot, original call, and report text.

---

Test paired categorical measurements

Description

Test paired categorical measurements

Usage

test_paired_categorical(
  data,
  before,
  after,
  alpha = 0.05,
  plot = TRUE,
  na.rm = TRUE
)

Arguments

data	A data frame.
before	Before categorical column.
after	After categorical column.
alpha	Significance level.
plot	Logical; include a ggplot object.
na.rm	Logical; remove missing values.

Value

A testflow object with class testflow_paired_categorical. The object is a list containing the cleaned paired categorical data, categorical descriptives, assumption checks, McNemar test result, discordant-pair table, optional ggplot, original call, and report text.

---

Test a one-sample proportion

Description

Test a one-sample proportion

Usage

test_proportion(
  data,
  outcome,
  success,
  p = 0.5,
  alpha = 0.05,
  plot = TRUE,
  na.rm = TRUE
)

Arguments

data	A data frame.
outcome	Categorical outcome column.
success	Value counted as success.
p	Reference probability.
alpha	Significance level.
plot	Logical; include a ggplot object.
na.rm	Logical; remove missing values.

Value

A testflow object with class testflow_proportion. The object is a list containing the cleaned data, categorical descriptives, assumption checks, recommended exact or approximate one-sample proportion test, primary test result with null hypothesis, alternative exact and approximate results, observed proportion summary, optional ggplot, original call, and report text.

References

Clopper, C. J., & Pearson, E. S. (1934). The use of confidence or fiducial limits illustrated in the case of the binomial. Biometrika, 26(4), 404-413.

---

Run a repeated-measures workflow from wide data

Description

Run a repeated-measures workflow from wide data

Usage

test_repeated(
  data,
  measures,
  id = NULL,
  between = NULL,
  alpha = 0.05,
  plot = TRUE,
  na.rm = TRUE
)

Arguments

data	A data frame.
measures	Repeated numeric columns selected with tidyselect syntax.
id	Optional subject identifier.
between	Optional between-subject factor.
alpha	Significance level.
plot	Logical; include a ggplot object.
na.rm	Logical; remove missing values.

Value

A testflow object with class testflow_repeated. The object is a list containing long-form repeated-measures data, numeric descriptives, assumption checks, recommended repeated-measures ANOVA or Friedman test, primary test result with null hypothesis, alternative repeated-measures results, post-hoc paired comparisons, effect size, optional ggplot, original call, and report text.

References

Fisher, R. A. (1925). Statistical Methods for Research Workers. Oliver and Boyd.

Friedman, M. (1937). The use of ranks to avoid the assumption of normality implicit in the analysis of variance. Journal of the American Statistical Association, 32(200), 675-701.

Wilcoxon, F. (1945). Individual comparisons by ranking methods. Biometrics Bulletin, 1(6), 80-83.

Girden, E. R. (1992). ANOVA: Repeated Measures. Sage.

---

Test repeated categorical measurements

Description

Test repeated categorical measurements

Usage

test_repeated_categorical(
  data,
  measures,
  id = NULL,
  alpha = 0.05,
  plot = TRUE,
  na.rm = TRUE
)

Arguments

data	A data frame.
measures	Repeated binary columns selected with tidyselect syntax.
id	Optional subject identifier.
alpha	Significance level.
plot	Logical; include a ggplot object.
na.rm	Logical; remove missing values.

Value

A testflow object with class testflow_repeated_categorical. The object is a list containing the cleaned repeated categorical data, descriptive counts, assumption checks, Cochran Q test result with null hypothesis, pairwise McNemar post-hoc comparisons, effect size, optional ggplot, original call, and report text.

References

Cochran, W. G. (1950). The comparison of percentages in matched samples. Biometrika, 37(3/4), 256-266.

McNemar, Q. (1947). Note on the sampling error of the difference between correlated proportions or percentages. Psychometrika, 12(2), 153-157.

---

Run a repeated-measures workflow from long data

Description

Run a repeated-measures workflow from long data

Usage

test_repeated_long(
  data,
  outcome,
  within,
  id,
  between = NULL,
  alpha = 0.05,
  plot = TRUE,
  na.rm = TRUE
)

Arguments

data	A data frame.
outcome	Numeric outcome column.
within	Within-subject time/condition column.
id	Subject identifier column.
between	Optional between-subject factor.
alpha	Significance level.
plot	Logical; include a ggplot object.
na.rm	Logical; remove missing values.

Value

A testflow object with class testflow_repeated. The object is a list containing the cleaned long-format data, numeric descriptives, assumption checks, recommended repeated-measures ANOVA or Friedman test, primary test result with null hypothesis, alternative repeated-measures results, post-hoc paired comparisons, effect size, optional ggplot, original call, and report text.

---

Compare a numeric outcome between two independent groups

Description

Compare a numeric outcome between two independent groups

Usage

test_two_groups(
  formula,
  data,
  group = NULL,
  alternative = c("two.sided", "less", "greater"),
  alpha = 0.05,
  plot = TRUE,
  na.rm = TRUE
)

Arguments

formula	A formula such as outcome ~ group, or a data frame when using pipe/data-first style.
data	A data frame, or an outcome column when using data-first style.
group	Two-level grouping column. Optional when using formula style.
alternative	Alternative hypothesis.
alpha	Significance level.
plot	Logical; include a ggplot object.
na.rm	Logical; remove missing values.

Value

A testflow object with class testflow_two_groups. The object is a list containing the cleaned data, descriptive statistics by group, assumption checks, recommended test, primary test result with null hypothesis, alternative test results, effect size, optional ggplot, original call, and report text.

---