spicy is an R package for frequency tables, cross-tabulations, association measures, categorical and continuous summary tables, publication-ready regression tables for 30+ model classes, and labelled survey data workflows.
Features
Every spicy table prints as readable text in the console and exports to gt, tinytable, flextable, Excel, Word, or the clipboard, following APA conventions. Around the tables, spicy provides the survey-data tools that feed them: variable inspection, codebooks, label extraction, and row-wise summaries.
-
Frequency tables with
freq(). -
Cross-tabulations with
cross_tab(): percentages, weights, chi-squared tests, and effect sizes. -
Association measures:
cramer_v(),phi(),gamma_gk(),kendall_tau_b(),somers_d()and 6 other coefficients;assoc_measures()computes the full set at once. -
Categorical and continuous summary tables with
table_categorical()andtable_continuous(), overall or by group. -
Model-based continuous summary tables with
table_continuous_lm()for linear-model reporting: classical / HC* / cluster-robust / bootstrap / jackknife variance, four effect-size families (f², Cohen’s d, Hedges’ g, Hays’ omega²) with noncentral CIs, additive covariate adjustment with G-computation (Statamarginsstyle) or equal-weight (emmeansstyle) marginal means, and weighted comparisons. -
Regression tables with
table_regression()for one or more fitted models side by side, across 30+ model classes (see Supported models below): classical / heteroskedasticity-robust / cluster-robust / bootstrap / jackknife variance with each class’s field-standard backend, standardised coefficients, family-awareexponentiate(OR / IRR / HR / RR / MR, link-gated), Wald or profile-likelihood CIs, average marginal effects (per-category for ordinal and multinomial models), partial f² / η² / ω² / χ² effect sizes, class-aware fit statistics (pseudo-R², Nakagawa marginal / conditional R², ICC), hierarchical model comparisons with the correct nested test per class, and multiple-comparison adjustment. Mixed models report their random effects as table rows with an optional boundary-correct per-term test; ordinal models report their thresholds; zero-inflated and hurdle models report every model component. Survival models go beyond hazard ratios: adjusted restricted-mean-survival-time and risk differences by g-computation (tau/at_time), for Cox fits (stratified included) and parametric AFT fits, in single tables and in the univariable screen. -
Univariable screening with
table_regression_uv(): one-predictor-at-a-time models merged with the multivariable fit, per-predictor N and events, forglm,lm, and Cox outcomes. -
Variable inspection with
varlist()andvl(): names, labels, values, classes, distinct values (N_distinct), valid observations (N_valid), and missing data. -
Codebooks with
code_book(): interactive and exportable, for labelled and survey-style datasets. -
Label extraction with
label_from_names(), including LimeSurvey-style headers. -
Row-wise summaries with
mean_n(),sum_n(), andcount_n(), with explicit control over missing values.
Works with labelled, factor, ordered, Date, POSIXct, and other common variable types – and honors declared missing values (haven’s na_values/na_range, tagged NAs) across the descriptive functions, with disclosure notes and a user_na escape hatch. For a full introduction, see Getting started with spicy.
Supported models
table_regression() accepts a single fit or a list of fits from any of these classes, and renders them with the conventions of each model family:
Class-specific structure renders as labelled blocks in the same table: random effects (with SE and CI on each variance component, and an optional boundary-correct per-term likelihood-ratio test), ordinal thresholds, non-proportional effects, zero-inflation and dispersion components. Robust variance requests use each class’s field-standard backend and are refused with a clear message when no valid backend exists.
The full capability map – each family’s marginal-effects estimand, exponentiate semantics, robust-variance backends and standardized-coefficient support – is the Supported models article; table_regression_models() returns the same registry as a data frame.
Installation
Install the current CRAN release, recommended for most users:
install.packages("spicy")Install the latest r-universe build:
install.packages(
"spicy",
repos = c(
"https://amaltawfik.r-universe.dev",
"https://cloud.r-project.org"
)
)This installs spicy from r-universe when available; CRAN is included only as a fallback for dependencies. The r-universe build may be newer than the current CRAN release.
Install the development version from GitHub with pak:
# install.packages("pak")
pak::pak("amaltawfik/spicy")Or with remotes:
# install.packages("remotes")
remotes::install_github("amaltawfik/spicy")Quick tour
The examples below use the bundled sochealth dataset.
Inspect variables
varlist(sochealth, tbl = TRUE)
#> # A tibble: 24 × 7
#> Variable Label Values Class N_distinct N_valid NAs
#> <chr> <chr> <chr> <chr> <int> <int> <int>
#> 1 sex Sex Femal… fact… 2 1200 0
#> 2 age Age (years) 25, 2… nume… 51 1200 0
#> 3 age_group Age group 25-34… orde… 4 1200 0
#> 4 education Highest education le… Lower… orde… 3 1200 0
#> 5 social_class Subjective social cl… Lower… orde… 5 1200 0
#> 6 region Region of residence Centr… fact… 6 1200 0
#> 7 employment_status Employment status Emplo… fact… 4 1200 0
#> 8 income_group Household income gro… Low, … orde… 4 1182 18
#> 9 income Monthly household in… 1000,… nume… 1052 1200 0
#> 10 smoking Current smoker No, Y… fact… 2 1175 25
#> # ℹ 14 more rows
code_book(
sochealth,
starts_with("bmi"),
values = TRUE,
include_na = TRUE
)See Explore variables and build codebooks for more on varlist(), vl(), and code_book().
Frequency tables and cross-tabulations
freq(sochealth, income_group)
#> Frequency table: income_group
#>
#> Category │ Values Freq. Percent Valid Percent
#> ────────────┼─────────────────────────────────────────────────────
#> Valid │ Low 247 20.6 20.9
#> │ Lower middle 388 32.3 32.8
#> │ Upper middle 328 27.3 27.7
#> │ High 219 18.2 18.5
#> Missing │ NA 18 1.5
#> ────────────┼─────────────────────────────────────────────────────
#> Total │ 1200 100.0 100.0
#>
#> Label: Household income group
#> Class: ordered, factor
#> Data: sochealth
cross_tab(sochealth, smoking, education, percent = "col")
#> Crosstable: smoking x education (Column %)
#>
#> Values │ Lower secondary Upper secondary Tertiary │ Total
#> ──────────┼──────────────────────────────────────────────────┼─────────
#> No │ 69.6 78.7 84.9 │ 78.8
#> Yes │ 30.4 21.3 15.1 │ 21.2
#> ──────────┼──────────────────────────────────────────────────┼─────────
#> Total │ 100.0 100.0 100.0 │ 100.0
#> N │ 257 527 391 │ 1175
#>
#> Chi-2(2) = 21.6, p <.001
#> Cramer's V = 0.14
#> Missing values removed: smoking (25).See Frequency tables and cross-tabulations for freq(), cross_tab(), percentages, weights, and tests.
Association measures
tbl <- xtabs(~ self_rated_health + education, data = sochealth)
# Quick scalar estimate
cramer_v(tbl)
#> [1] 0.1761697
# Detailed result with CI and p-value
cramer_v(tbl, detail = TRUE)
#> Estimate SE CI lower CI upper p
#> 0.176 -- 0.120 0.231 <.001See Cramer’s V, Phi, and association measures for a guide on choosing the right measure.
Summary tables
table_categorical(
sochealth,
select = c(smoking, physical_activity),
labels = c(
smoking = "Current smoker",
physical_activity = "Physical activity"
)
)
#> Categorical table
#>
#> Variable │ n %
#> ─────────────────────┼───────────────
#> Current smoker │
#> No │ 926 77.2
#> Yes │ 249 20.8
#> (Missing) │ 25 2.1
#> ╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
#> Physical activity │
#> No │ 650 54.2
#> Yes │ 550 45.8
table_categorical(
sochealth,
select = c(smoking, physical_activity),
by = education,
labels = c(
smoking = "Current smoker",
physical_activity = "Physical activity"
)
)
#> Categorical table by education
#>
#> Variable │ Lower secondary n Lower secondary % Upper secondary n
#> ───────────────────┼─────────────────────────────────────────────────────────
#> Current smoker │
#> No │ 179 68.6 415
#> Yes │ 78 29.9 112
#> (Missing) │ 4 1.5 12
#> ╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
#> Physical activity │
#> No │ 177 67.8 310
#> Yes │ 84 32.2 229
#>
#> Variable │ Upper secondary % Tertiary n Tertiary % Total n
#> ───────────────────┼────────────────────────────────────────────────────
#> Current smoker │
#> No │ 77.0 332 83.0 926
#> Yes │ 20.8 59 14.8 249
#> (Missing) │ 2.2 9 2.2 25
#> ╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
#> Physical activity │
#> No │ 57.5 163 40.8 650
#> Yes │ 42.5 237 59.2 550
#>
#> Variable │ Total % p Cramer's V
#> ───────────────────┼────────────────────────────
#> Current smoker │ <.001 .14
#> No │ 77.2
#> Yes │ 20.8
#> (Missing) │ 2.1
#> ╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
#> Physical activity │ <.001 .21
#> No │ 54.2
#> Yes │ 45.8
table_continuous(
sochealth,
select = c(bmi, life_sat_health)
)
#> Descriptive statistics
#>
#> Variable │ M SD Min Max 95% CI LL
#> ────────────────────────────────┼──────────────────────────────────────
#> Body mass index │ 25.93 3.72 16.00 38.90 25.72
#> Satisfaction with health (1-5) │ 3.55 1.25 1.00 5.00 3.48
#>
#> Variable │ 95% CI UL n
#> ────────────────────────────────┼─────────────────
#> Body mass index │ 26.14 1188
#> Satisfaction with health (1-5) │ 3.62 1192
#>
#> Missing values removed: bmi (12), life_sat_health (8).
table_continuous(
sochealth,
select = c(bmi, life_sat_health),
by = education
)
#> Descriptive statistics
#>
#> Variable │ Group M SD Min Max
#> ────────────────────────────────┼────────────────────────────────────────────
#> Body mass index │ Lower secondary 28.09 3.47 18.20 38.90
#> │ Upper secondary 26.02 3.43 16.00 37.10
#> │ Tertiary 24.39 3.52 16.00 33.00
#> ╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
#> Satisfaction with health (1-5) │ Lower secondary 2.71 1.20 1.00 5.00
#> │ Upper secondary 3.53 1.19 1.00 5.00
#> │ Tertiary 4.11 1.04 1.00 5.00
#>
#> Variable │ Group 95% CI LL 95% CI UL n
#> ────────────────────────────────┼────────────────────────────────────────────
#> Body mass index │ Lower secondary 27.66 28.51 260
#> │ Upper secondary 25.73 26.31 534
#> │ Tertiary 24.04 24.74 394
#> ╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
#> Satisfaction with health (1-5) │ Lower secondary 2.57 2.86 259
#> │ Upper secondary 3.43 3.63 534
#> │ Tertiary 4.01 4.21 399
#>
#> Variable │ Group p
#> ────────────────────────────────┼────────────────────────
#> Body mass index │ Lower secondary <.001
#> │ Upper secondary
#> │ Tertiary
#> ╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
#> Satisfaction with health (1-5) │ Lower secondary <.001
#> │ Upper secondary
#> │ Tertiary
#>
#> Missing values removed: bmi (12), life_sat_health (8).
table_continuous_lm(
sochealth,
select = c(wellbeing_score, bmi),
by = sex,
vcov = "HC3",
output = "data.frame"
)
#> Variable M (Female) M (Male)
#> wellbeing_score WHO-5 wellbeing index (0-100) 67.16194 71.04879
#> bmi Body mass index 25.68506 26.19685
#> Δ (Male - Female) 95% CI LL 95% CI UL p R²
#> wellbeing_score 3.8868576 2.12265210 5.6510631 1.670572e-05 0.015475137
#> bmi 0.5117882 0.08904596 0.9345305 1.769614e-02 0.004728908
#> n
#> wellbeing_score 1200
#> bmi 1188
fit <- lm(wellbeing_score ~ age + sex + smoking, data = sochealth)
table_regression(fit)
#> Linear regression: wellbeing_score
#>
#> Variable │ B SE 95% CI p
#> ─────────────────┼──────────────────────────────────────
#> (Intercept) │ 65.20 1.66 [61.95, 68.45] <.001
#> age │ 0.05 0.03 [-0.01, 0.11] .130
#> sex: │
#> Female (ref.) │ – – – –
#> Male │ 3.86 0.91 [ 2.08, 5.63] <.001
#> smoking: │
#> No (ref.) │ – – – –
#> Yes │ -1.72 1.11 [-3.89, 0.45] .121
#> ╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
#> n │ 1175
#> R² │ 0.02
#> Adj.R² │ 0.02
#>
#> Note. Linear regression.
#> Std. errors: classical (OLS).Regression tables cover 30+ model classes with the conventions of each family. A mixed-effects fit, for example, reports its random effects as rows (SD, correlation, residual – each with SE and CI), the ICC and group sizes as fit statistics, and the likelihood-ratio test of the random part with the boundary-correct chi-bar-squared p-value:
library(lme4)
fit_mixed <- lmer(Reaction ~ Days + (Days | Subject), data = sleepstudy)
table_regression(fit_mixed)
#> Linear mixed-effects regression: Reaction
#>
#> Variable │ B SE 95% CI p
#> ─────────────────────────────────┼────────────────────────────────────────
#> (Intercept) │ 251.41 6.82 [238.03, 264.78] <.001
#> Days │ 10.47 1.55 [ 7.44, 13.50] <.001
#> ╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
#> Random effects: │
#> σ Subject (Intercept) │ 24.74 5.84 [ 6.79, 34.32] –
#> σ Subject Days │ 5.92 1.25 [ 2.47, 8.00] –
#> ρ Subject ((Intercept), Days) │ 0.07 0.33 [ -0.57, 0.70] –
#> σ (Residual) │ 25.59 1.51 [ 22.44, 28.39] –
#> ╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
#> n │ 180
#> N (Subject) │ 18
#> R² (marginal) │ 0.28
#> R² (conditional) │ 0.80
#> AIC │ 1755.6
#> BIC │ 1774.8
#>
#> Note. Linear mixed-effects regression.
#> Std. errors: Wald (model-based).
#> p-values: Wald-z, large-sample approximation. Load `lmerTest` for Satterthwaite t-tests.
#> Random effects (REML): LR test vs linear regression, χ̄²(3) = 150.04, p < .001.See Categorical summary tables for categorical summaries, Continuous summary tables for continuous summaries and group comparisons, Model-based continuous summary tables for weighted or robust linear-model reporting, Regression coefficient tables for regression tables across model families, and Summary tables for APA-style reporting for an overview of summary tables.
Row-wise summaries
df <- data.frame(
x1 = c(10, NA, 30, 40, 50),
x2 = c(5, NA, 15, NA, 25),
x3 = c(NA, 30, 20, 50, 10)
)
mean_n(df)
#> [1] NA NA 21.66667 NA 28.33333
sum_n(df, min_valid = 2)
#> [1] 15 NA 65 90 85
count_n(df, special = "NA")
#> [1] 1 2 0 1 0See Getting started with spicy for a longer workflow using mean_n(), sum_n(), and count_n().
Label extraction
# LimeSurvey-style headers: "code. label"
df <- tibble::tibble(
"age. Age of respondent" = c(25, 30),
"score. Total score" = c(12, 14)
)
out <- label_from_names(df)
labelled::var_label(out)
#> $age
#> [1] "Age of respondent"
#>
#> $score
#> [1] "Total score"See Explore variables and build codebooks for more on label_from_names(), varlist(), and code_book().
Documentation
Each workflow has a dedicated vignette:
- Getting started with spicy
- Explore variables and build codebooks
- Frequency tables and cross-tabulations
- Cramer’s V, Phi, and association measures
- Categorical summary tables
- Continuous summary tables
- Model-based continuous summary tables
- Regression coefficient tables
- Mixed-effects regression tables
- Count and two-part regression tables
- Survival regression tables
- Ordinal regression tables
- Multinomial regression tables
- Summary tables for APA-style reporting
Key reference pages: freq(), cross_tab(), cramer_v(), table_categorical(), table_continuous(), table_continuous_lm(), table_regression(), varlist(), code_book(), label_from_names(), mean_n(), sum_n(), and count_n() – see the full function index for everything else.
Citation
To cite spicy in a publication or teaching material (this entry is generated by citation("spicy") at build time and always refers to the current CRAN release):
Tawfik A (2026). spicy: Descriptive Statistics, Summary Tables, and Data Management Tools. doi:10.32614/CRAN.package.spicy https://doi.org/10.32614/CRAN.package.spicy. R package version 0.12.0, https://CRAN.R-project.org/package=spicy.
@Manual{,
title = {spicy: Descriptive Statistics, Summary Tables, and Data Management Tools},
author = {Amal Tawfik},
year = {2026},
note = {R package version 0.12.0},
url = {https://CRAN.R-project.org/package=spicy},
doi = {10.32614/CRAN.package.spicy},
}- Package DOI: https://doi.org/10.32614/CRAN.package.spicy.
- Source citation file: https://github.com/amaltawfik/spicy/blob/main/inst/CITATION
License
MIT. See LICENSE for details.