Package {ExpDesignR}

Type: Package
Title: Experimental Design and Randomization Methods for Biomedical and Veterinary Research
Version: 1.0.0
Description: Provides reproducible methods for experimental design and treatment allocation in biomedical, veterinary, agricultural, and clinical research. Includes simple, fixed-block, variable-block, stratified, stratified-block, cluster, matched-pair, restricted, minimization, and covariate-adaptive randomization, together with completely randomized, randomized-block, factorial, split-plot, Latin square, and crossover designs. Also provides allocation summaries, balance diagnostics, schedule export, and visualization. The methods are based on established principles of randomization and experimental design; see Rosenberger and Lachin (2015, ISBN:9781118742242) and Jones and Kenward (2014, ISBN:9781439861424).
License: MIT + file LICENSE
Encoding: UTF-8
Depends: R (≥ 4.2.0)
Imports: dplyr, ggplot2, rlang, stats, tibble, utils
Suggests: covr, knitr, rmarkdown, spelling, testthat (≥ 3.0.0)
URL: https://github.com/vinodhpmd/ExpDesignR
BugReports: https://github.com/vinodhpmd/ExpDesignR/issues
Language: en-US
Config/testthat/edition: 3
Config/roxygen2/version: 8.1.0
VignetteBuilder: knitr
NeedsCompilation: no
Packaged: 2026-08-30 07:06:01 UTC; m
Author: Vinodhkumar Obli Rajendran [aut, cre], Keerthi Aaradhana [aut]
Maintainer: Vinodhkumar Obli Rajendran <vinodhkumar.rajendran@gmail.com>
Repository: CRAN
Date/Publication: 2026-09-01 12:40:08 UTC

Allocation Summary

Description

Summarizes treatment allocations from a randomization schedule.

Usage

allocation_summary(schedule, group_col = "Group")

Arguments

schedule

A data frame or tibble produced by an ExpDesignR randomization function.

group_col

Name of the treatment group column.

Value

A tibble summarizing the number and percentage of subjects in each treatment group.

Examples

sch <- simple_randomization(
  n = 20,
  groups = c("Control","Treatment"),
  seed = 123
)

allocation_summary(sch)

Check Treatment Allocation Balance

Description

Summarize treatment-count and percentage imbalance in an allocation schedule.

Usage

balance_check(schedule, group_col = "Group")

Arguments

schedule

A data frame or tibble containing treatment assignments.

group_col

Name of the treatment column.

Value

A tibble with treatment counts, percentages, and balance statistics.

Examples

x <- block_randomization(40, c("A", "B"), 4, seed = 1)
balance_check(x)

Fixed Block Randomization

Description

Generate a balanced fixed-block randomization schedule.

Usage

block_randomization(n, groups, block_size = 4, seed = NULL, ratio = NULL)

Arguments

n

Number of subjects. It must be divisible by 'block_size'.

groups

Character vector of treatment groups.

block_size

Size of each block.

seed

Optional random seed.

ratio

Optional allocation weights.

Value

A tibble with subject, block, and treatment assignment.

Examples

block_randomization(
  24,
  c("Control", "Treatment"),
  4,
  seed = 123
)

Cluster Randomization

Description

Randomly assigns intact clusters to treatment groups.

Usage

cluster_randomization(clusters, groups, seed = NULL, ratio = NULL)

Arguments

clusters

Character or numeric vector of unique cluster IDs.

groups

Treatment groups.

seed

Optional random seed.

ratio

Optional allocation weights.

Value

A tibble containing cluster assignments.

Examples

cluster_randomization(paste0("Farm_", 1:20), c("Control", "Treatment"), seed = 123)

Completely Randomized Design

Description

Randomly assign experimental units to treatment groups.

Usage

completely_randomized_design(n, treatments, seed = NULL, ratio = NULL)

Arguments

n

Number of units.

treatments

Treatment labels.

seed

Optional random seed.

ratio

Optional allocation weights.

Value

A tibble containing unit and treatment.

Covariate-Adaptive Randomization

Description

Allocate subjects sequentially while reducing imbalance across categorical covariates. A probabilistic element preserves allocation randomness.

Usage

covariate_adaptive_randomization(
  data,
  covariates,
  groups = c("Control", "Treatment"),
  p = 0.75,
  seed = NULL
)

Arguments

data

Study-subject data frame.

covariates

Covariate column names.

groups

Treatment groups.

p

Probability of selecting one of the best-scoring groups.

seed

Optional random seed.

Value

A tibble containing the original data and 'Treatment'.

Examples

dat <- data.frame(ID = 1:20, Sex = rep(c("M", "F"), 10), Site = rep(c("A", "B"), 10))
covariate_adaptive_randomization(dat, c("Sex", "Site"), seed = 1)

Crossover Design

Description

Generates a crossover design for clinical, veterinary, pharmaceutical and agricultural experiments.

Usage

crossover_design(
  treatments,
  subjects,
  periods = length(treatments),
  seed = NULL
)

Arguments

treatments

Character vector of treatment labels.

subjects

Number of subjects.

periods

Number of study periods.

seed

Optional random seed.

Value

A tibble containing the crossover schedule.

Examples

crossover_design(
  treatments = c("A","B"),
  subjects = 8,
  periods = 2,
  seed = 123
)

Export Randomization Schedule

Description

Export a randomization schedule to a CSV file.

Usage

export_schedule(schedule, file, row.names = FALSE)

Arguments

schedule

A data frame or tibble generated by ExpDesignR.

file

Character. Output CSV filename or path. This argument must be supplied explicitly.

row.names

Logical. Should row names be written?

Value

Invisibly returns the input schedule unchanged. The function writes the schedule to the CSV file specified by file.

Examples

sch <- simple_randomization(
  n = 20,
  groups = c("Control", "Treatment"),
  seed = 123
)

tf <- tempfile(fileext = ".csv")

export_schedule(
  sch,
  file = tf
)

unlink(tf)

Factorial Design

Description

Generate a randomized full-factorial treatment combination design.

Usage

factorial_design(factors, replicates = 1L, seed = NULL)

Arguments

factors

Named list of factor levels.

replicates

Number of replicates per combination.

seed

Optional random seed.

Value

A tibble containing randomized factorial combinations.

Latin Square Design

Description

Generates a Latin Square design for experimental studies.

Usage

latin_square(treatments, randomize = TRUE, seed = NULL)

Arguments

treatments

Character vector of treatment labels.

randomize

Logical. Should rows, columns and treatments be randomized? Default is TRUE.

seed

Optional random seed.

Value

A matrix representing a Latin square.

Examples

latin_square(
  treatments = LETTERS[1:4],
  seed = 123
)

Matched-Pair Randomization

Description

Randomly assigns one member of each matched pair to each of two treatments.

Usage

matched_pair_randomization(
  data,
  pair,
  groups = c("Control", "Treatment"),
  seed = NULL
)

Arguments

data

Study-subject data frame.

pair

Column containing matched-pair IDs.

groups

Exactly two treatment labels.

seed

Optional random seed.

Value

A tibble containing the original data and treatment assignment.

Examples

dat <- data.frame(
  ID = 1:10,
  Pair = rep(1:5, each = 2)
)

matched_pair_randomization(
  dat,
  "Pair",
  c("Control", "Treatment"),
  seed = 1
)

Minimization Randomization

Description

Perform covariate-adaptive minimization by selecting the treatment that gives the smallest resulting marginal imbalance, with optional randomness.

Usage

minimization_randomization(
  data,
  covariates,
  groups = c("Control", "Treatment"),
  probability = 0.8,
  seed = NULL
)

Arguments

data

Study-subject data frame.

covariates

Character vector of categorical covariate columns.

groups

Treatment groups.

probability

Probability of selecting a best-scoring group.

seed

Optional random seed.

Value

A tibble containing the original data and treatment allocation.

Examples

dat <- data.frame(ID = 1:30, Sex = rep(c("M", "F"), 15), Site = rep(LETTERS[1:3], 10))
minimization_randomization(dat, c("Sex", "Site"), seed = 123)

Plot Randomization Schedule

Description

Creates a bar chart showing the number of subjects allocated to each treatment group.

Usage

plot_randomization(
  schedule,
  group_col = "Group",
  fill = "#2C7FB8",
  title = "Treatment Allocation"
)

Arguments

schedule

A data frame produced by ExpDesignR.

group_col

Character. Name of the treatment column.

fill

Character. Fill colour.

title

Character. Plot title.

Value

A ggplot object.

Examples

sch <- simple_randomization(
  n = 40,
  groups = c("Control","Treatment"),
  seed = 123
)

plot_randomization(sch)

Randomization Diagnostics

Description

Return compact diagnostics for a treatment allocation schedule.

Usage

randomization_diagnostics(schedule, group_col = "Group")

Arguments

schedule

A data frame or tibble containing treatment assignments.

group_col

Name of the treatment column.

Value

A named list of allocation diagnostics.

Examples

x <- simple_randomization(50, c("A", "B"), seed = 1)
randomization_diagnostics(x)

Randomized Block Design

Description

Randomize units within balanced blocks.

Usage

randomized_block_design(n, treatments, block_size = 4, seed = NULL)

Arguments

n

Number of units.

treatments

Treatment labels.

block_size

Block size.

seed

Optional random seed.

Value

A tibble with unit, block, and treatment.

Restricted Randomization

Description

Generate simple random allocations subject to a maximum treatment-count imbalance.

Usage

restricted_randomization(
  n,
  groups,
  max_imbalance = 1,
  seed = NULL,
  ratio = NULL
)

Arguments

n

Number of subjects.

groups

Treatment groups.

max_imbalance

Maximum allowed difference between the largest and smallest group counts.

seed

Optional random seed.

ratio

Optional allocation weights.

Value

A tibble containing the restricted allocation.

Examples

restricted_randomization(30, c("A", "B"), max_imbalance = 2, seed = 1)

Simple Randomization

Description

Generate a simple random allocation schedule.

Usage

simple_randomization(n, groups, seed = NULL, ratio = NULL)

Arguments

n

Number of subjects.

groups

Character vector of treatment groups.

seed

Optional random seed.

ratio

Optional positive allocation weights for the groups.

Value

A tibble containing subject IDs and assigned groups.

Examples

simple_randomization(20, c("Control", "Treatment"), seed = 123)

Split-Plot Design

Description

Generate a randomized split-plot treatment schedule.

Usage

split_plot_design(whole, sub, n_whole = length(whole), seed = NULL)

Arguments

whole

Whole-plot treatment labels.

sub

Sub-plot treatment labels.

n_whole

Number of whole plots.

seed

Optional random seed.

Value

A tibble with whole plots and sub-plots.

Stratified Block Randomization

Description

Perform blocked randomization independently within each stratum.

Usage

stratified_block_randomization(
  data,
  strata,
  groups,
  block_size = 4,
  seed = NULL,
  ratio = NULL
)

Arguments

data

Study-subject data frame.

strata

Character vector of stratification columns.

groups

Treatment groups.

block_size

Block size.

seed

Optional random seed.

ratio

Optional allocation weights.

Value

A tibble containing the original data, stratum, block, and treatment.

Examples

dat <- data.frame(ID = 1:40, Sex = rep(c("M", "F"), each = 20))
stratified_block_randomization(dat, "Sex", c("A", "B"), 4, seed = 1)

Stratified Randomization

Description

Randomize independently within one or more strata.

Usage

stratified_randomization(data, strata, groups, seed = NULL, ratio = NULL)

Arguments

data

Study-subject data frame.

strata

Character vector of stratification columns.

groups

Treatment groups.

seed

Optional random seed.

ratio

Optional allocation weights.

Value

The input data with a 'Treatment' column.

Examples

dat <- data.frame(ID = 1:20, Sex = rep(c("M", "F"), each = 10))
stratified_randomization(dat, "Sex", c("Control", "Treatment"), seed = 123)

Variable Block Randomization

Description

Generate randomization using randomly selected permitted block sizes.

Usage

variable_block_randomization(
  n,
  groups,
  block_sizes = c(4, 6, 8),
  seed = NULL,
  ratio = NULL
)

Arguments

n

Number of subjects.

groups

Treatment groups.

block_sizes

Permitted block sizes. Each must support the requested allocation ratio.

seed

Optional random seed.

ratio

Optional allocation weights.

Value

A tibble with subject, block, block size, and group.

Examples


variable_block_randomization(
  30,
  c("Control", "Treatment"),
  c(4, 6, 8),
  seed = 123
)