spicy: frequency tables, cross-tabulations, summary and regression tables in R

spicy is an R package for publication-ready tables: frequency tables and cross-tabulations, association measures, categorical and continuous summary tables, regression tables for 39 model classes, and labelled survey data workflows.

Features

Every summary and regression table prints as readable text in the console and exports to gt, tinytable, flextable, Excel, Word, or the clipboard, following rigorous reporting conventions – APA by default, with named journal styles. 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() and table_continuous(), overall or by group.
Multiple-response tables with table_multiresponse(): “check all that apply” items read as one question, overall or by group, with a global test that allows for several answers per respondent. The function is experimental in this release.
Outcome-first summary tables with table_outcome(): one continuous outcome described across several groupings, one block of rows per grouping variable, a group comparison per block, and a marginal Overall row – the transpose of table_continuous(), with the same statistic vocabulary and output engines.
Design-based summary tables with table_continuous_svy() and table_categorical_svy() on a survey::svydesign object: estimates, SEs and CIs from the survey package (Taylor linearization or replicate weights), design degrees of freedom, observed and weighted counts – the design twins of the two families above, with the same statistic vocabulary and output engines.
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’s g, Hays’s omega²) with noncentral CIs, additive covariate adjustment with G-computation (Stata margins style) or equal-weight (emmeans style) marginal means, and weighted comparisons.
Regression tables with table_regression() for one or more fitted models side by side, across 39 model classes in 12 families (see Supported models below): classical / heteroskedasticity-robust / cluster-robust / bootstrap / jackknife variance with each class’s field-standard backend, standardized coefficients, family-aware exponentiate (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 (a Wald test on the requested matrix under a robust or cluster-robust vcov), 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, and cumulative-link mixed models (ordinal::clmm(), glmmTMB with the ordinal family) both; 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 coxph fits (stratified included) and survreg 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, for glm, lm, and Cox outcomes.
Journal styles with spicy_style(): style = "jama", "nejm", "lancet", "annals", "apa", or "aer" restyles any table, every rule sourced from each style’s published guidelines. For a French table, options(spicy.language = "fr") gives the words and the French typography together.
Variable inspection with varlist() and vl(): names, labels, values, classes, distinct values (N_distinct), valid observations (N_valid), and missing data.
Codebooks with code_book(): variables, values, and unweighted counts, printed in the console, written to Excel, or compiled to a PDF with a cover, notes (paragraphs, lists, italics, bold, code, links), one sheet per variable, and an index.
Label extraction with label_from_names(), including LimeSurvey-style headers.
Row-wise summaries with mean_n(), sum_n(), and count_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.

What makes the tables dependable

One table, every output, verified. The console rendering and the gt, tinytable, flextable, Word, Excel, and clipboard outputs are kept cell-for-cell identical by dedicated parity test batteries – significance stars, events/N counts, reference rows, block structure, captions, and disclosure notes included. What you publish is what you saw.
Numbers validated against reference implementations. Estimates, standard errors, and association measures are pinned, to the digit, against independent software and packages (SPSS/PSPP conventions for contingency measures, Stata conventions for cluster-robust variance, survival::concordance(), exact order-statistic confidence intervals for medians), and each convention is documented where the number is produced.
Disclosure by default. Whatever affects the numbers is stated in the table’s own notes, on every output: removed missing values, estimation samples, standard-error types, test choices, multiple-comparison families.
A typed, versioned table contract. as_structured() returns the complete typed view of a regression or summary table – values, precisions, significance markers, row roles – with an explicit contract version, so custom renderers and downstream pipelines can detect changes instead of silently mis-reading them.
Hard errors over silent guesses. Requests a model class cannot honor are refused with a classed, actionable error – never rendered as an empty column or a silently substituted statistic.

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:

Family	Engines
Linear and generalized linear	`stats::lm()`, `stats::glm()`, `MASS::glm.nb()`, `MASS::rlm()`, `stats::nls()`
Robust, IV, quantile, panel	`estimatr::lm_robust()`, `estimatr::iv_robust()`, `AER::ivreg()`, `AER::tobit()`, `quantreg::rq()`, `fixest::feols()`, `fixest::feglm()`, `fixest::fepois()`, `fixest::fenegbin()`
Mixed effects	`lme4::lmer()`, `lme4::glmer()`, `glmmTMB::glmmTMB()`, `nlme::lme()`, `nlme::gls()`
Ordinal and categorical	`MASS::polr()`, `ordinal::clm()` (incl. partial proportional odds), `ordinal::clmm()`, `glmmTMB::glmmTMB()` (ordinal family), `nnet::multinom()`, `mlogit::mlogit()`
Counts, two-part models	`pscl::hurdle()`, `pscl::zeroinfl()`, `glmmTMB::glmmTMB()` (zero-inflation and dispersion components)
Survival	`survival::coxph()`, `survival::survreg()`, `rms::cph()`, `flexsurv::flexsurvreg()`
Survey-weighted	`survey::svyglm()`, `survey::svyolr()`, `survey::svycoxph()` (design-based SEs)
Population-averaged (GEE)	`geepack::geeglm()` (native sandwich SEs, working correlation disclosed)
Additive, proportions, selection	`mgcv::gam()`, `mgcv::bam()`, `betareg::betareg()`, `sampleSelection::selection()`
rms	`rms::ols()`, `rms::lrm()`, `rms::Glm()`
Bayesian	`rstanarm::stan_glm()`, `rstanarm::stan_glmer()`, `brms::brm()` (posterior median, credible intervals, no p-values)

Class-specific structure renders as labeled 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 demo with labelled data

varlist(sochealth, tbl = TRUE)
#> # A tibble: 29 × 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 19 more rows

Codebooks

code_book() documents a data frame: the list of variables in the console, an Excel workbook, or a PDF with a cover, a page about the data, one sheet per variable, and an index. The notes take paragraphs and list items, with italics, bold, code, and links.

code_book(
  sochealth,
  title = "Social health survey",
  subtitle = "Simulated data shipped with spicy",
  authors = list(
    list(
      name = "Jane Doe",
      affiliation = "University of Somewhere",
      orcid = "0000-0002-1825-0097"
    ),
    list(name = "John Doe", affiliation = "Somewhere Institute of Public Health")
  ),
  notes = c(
    "Simulated data shipped with *spicy* (`?sochealth`): 1,200 respondents of a fictitious social health survey.",
    "- `weight` is the survey design weight: the counts of this codebook are **unweighted**."
  ),
  output = "sochealth_codebook.pdf"
)

The cover of the PDF codebook: the word CODEBOOK in spaced capitals, the title Social health survey, its subtitle, two authors with their affiliations and an ORCID link, and the date

The sheet of one variable in the PDF codebook: a band with its position, name, and type, its label, the counts of valid, missing, and distinct values, and the table of its values with n, percent, and valid percent, the system missing row in gray

The whole document: the PDF, the same in French, and the Excel workbook. 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
#> Cramér'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  <.001

See Cramér’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    78.8  
#>    Yes               │  249    21.2  
#>    (Missing)         │   25          
#> ╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
#>  Physical activity   │               
#>    No                │  650    54.2  
#>    Yes               │  550    45.8  
#> 
#> Note. N = 1200.

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               69.6                415        
#>    Yes             │         78               30.4                112        
#>    (Missing)       │          4                                    12        
#> ╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
#>  Physical activity │                                                         
#>    No              │        177               67.8                310        
#>    Yes             │         84               32.2                229        
#> 
#>  Variable          │ Upper secondary %  Tertiary n  Tertiary %  Total n 
#> ───────────────────┼────────────────────────────────────────────────────
#>  Current smoker    │                                                    
#>    No              │       78.7            332         84.9       926   
#>    Yes             │       21.3             59         15.1       249   
#>    (Missing)       │                         9                     25   
#> ╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
#>  Physical activity │                                                    
#>    No              │       57.5            163         40.8       650   
#>    Yes             │       42.5            237         59.2       550   
#> 
#>  Variable          │ Total %    p     V  
#> ───────────────────┼─────────────────────
#>  Current smoker    │          <.001  .14 
#>    No              │  78.8               
#>    Yes             │  21.2               
#>    (Missing)       │                     
#> ╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
#>  Physical activity │          <.001  .21 
#>    No              │  54.2               
#>    Yes             │  45.8               
#> 
#> Note. N = 1200: Lower secondary (n = 261), Upper secondary (n = 539), Tertiary (n = 400). Percentages are column percentages. P-values: Pearson chi-squared test. V = Cramér's V.

table_continuous(
  sochealth,
  select = c(bmi, life_sat_health)
)
#> Descriptive statistics
#> 
#>  Variable        │   M     SD    Min    Max   95% CI LL  95% CI UL   n   
#> ─────────────────┼───────────────────────────────────────────────────────
#>  bmi             │ 25.93  3.72  16.00  38.90    25.72      26.14    1188 
#>  life_sat_health │  3.55  1.25   1.00   5.00     3.48       3.62    1192 
#> 
#> Note. N = 1200. Missing values removed: bmi (12), life_sat_health (8). CI = confidence interval; LL = lower limit; UL = upper limit.

table_continuous(
  sochealth,
  select = c(bmi, life_sat_health),
  by = education
)
#> Descriptive statistics by education
#> 
#>  Variable        │ Group              M     SD    Min    Max   95% CI LL 
#> ─────────────────┼───────────────────────────────────────────────────────
#>  bmi             │ Lower secondary  28.09  3.47  18.20  38.90    27.66   
#>                  │ Upper secondary  26.02  3.43  16.00  37.10    25.73   
#>                  │ Tertiary         24.39  3.52  16.00  33.00    24.04   
#> ╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
#>  life_sat_health │ Lower secondary   2.71  1.20   1.00   5.00     2.57   
#>                  │ Upper secondary   3.53  1.19   1.00   5.00     3.43   
#>                  │ Tertiary          4.11  1.04   1.00   5.00     4.01   
#> 
#>  Variable        │ Group            95% CI UL   n     p   
#> ─────────────────┼────────────────────────────────────────
#>  bmi             │ Lower secondary    28.51    260  <.001 
#>                  │ Upper secondary    26.31    534        
#>                  │ Tertiary           24.74    394        
#> ╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
#>  life_sat_health │ Lower secondary     2.86    259  <.001 
#>                  │ Upper secondary     3.63    534        
#>                  │ Tertiary            4.21    399        
#> 
#> Note. N = 1200: Lower secondary (n = 261), Upper secondary (n = 539), Tertiary (n = 400). Missing values removed: bmi (12), life_sat_health (8). P-values: Welch one-way ANOVA. CI = confidence interval; LL = lower limit; UL = upper limit.

table_continuous_lm(
  sochealth,
  select = c(wellbeing_score, bmi),
  by = sex,
  vcov = "HC3",
  output = "data.frame"
)
#>          variable           label M (Female) M (Male)         Δ  95% CI LL
#> 1 wellbeing_score wellbeing_score   67.16194 71.04879 3.8868576 2.12265210
#> 2             bmi             bmi   25.68506 26.19685 0.5117882 0.08904596
#>   95% CI UL            p          R²    n
#> 1 5.6510631 1.670572e-05 0.015475137 1200
#> 2 0.9345305 1.769614e-02 0.004728908 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). CI = confidence interval.

Regression tables cover 39 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. CI = confidence interval; AIC = Akaike information criterion; BIC = Bayesian information criterion.

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, Publication-ready regression tables for table_regression() across model families, and Summary tables for reporting for an overview of summary tables.

The same numbers can be laid out outcome-first: one outcome, several groupings, one block per grouping.

table_outcome(
  sochealth,
  outcome = bmi,
  select = c(sex, smoking)
)
#> Descriptive statistics of bmi
#> 
#>  Variable    │   M     SD    Min    Max   95% CI LL  95% CI UL   n     p   
#> ─────────────┼─────────────────────────────────────────────────────────────
#>  Overall     │ 25.93  3.72  16.00  38.90    25.72      26.14    1188       
#> ╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
#>  sex         │                                                        .018 
#>    Female    │ 25.69  3.78  16.00  38.90    25.39      25.98     616       
#>    Male      │ 26.20  3.64  16.00  37.70    25.90      26.50     572       
#> ╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
#>  smoking     │                                                        .903 
#>    No        │ 25.96  3.76  16.00  38.90    25.72      26.21     915       
#>    Yes       │ 25.93  3.58  16.80  35.30    25.48      26.38     248       
#>    (Missing) │ 24.74  3.63  17.60  32.50    23.24      26.23      25       
#> 
#> Note. N = 1200. Missing values removed: bmi (12). P-values: Welch t-test. Each block compares bmi across the levels of one variable, without controlling for the other variables. Overall = the whole analytic sample. CI = confidence interval; LL = lower limit; UL = upper limit.

A “check all that apply” question gets a table of its own, one row per item. In sochealth, five logical items record the sources of health information a respondent used in the past 12 months. The article Multiple-response questions explains the bases, the percentages and the tests.

table_multiresponse(
  sochealth,
  select = starts_with("info_"),
  by = sex
)
#> Multiple responses by sex
#> 
#>  Variable         │ Female n  Female %  Male n  Male %  Total n  Total %    p   
#> ──────────────────┼─────────────────────────────────────────────────────────────
#>  info_doctor      │   377       62.9     344     60.9     721     61.9    1.000 
#>  info_pharmacist  │   137       22.9     119     21.1     256     22.0    1.000 
#>  info_internet    │   268       44.7     255     45.1     523     44.9    1.000 
#>  info_family      │   219       36.6     141     25.0     360     30.9    <.001 
#>  info_media       │    78       13.0      63     11.2     141     12.1    1.000 
#>  (No item chosen) │    31        5.2      36      6.4      67      5.8     –    
#> ╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
#>  Base             │   599                565             1164                   
#> 
#> Note. Multiple answers possible: percentages can sum to more than 100%. Base: respondents who answered the question (excluded: 36 with no answer to the question). Item p-values: Pearson chi-squared test of each item by group. Adjusted for 5 tests (Holm method). Overall group comparison: Rao-Scott second-order corrected chi-squared test of multiple marginal independence, χ²(4.61) = 18.78, p = .002.

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 0

See 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 article on the package site:

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: Publication-Ready Tables for Descriptive Statistics and Regression Models. doi:10.32614/CRAN.package.spicy https://doi.org/10.32614/CRAN.package.spicy. R package version 0.14.0, https://CRAN.R-project.org/package=spicy.

@Manual{,
  title = {spicy: Publication-Ready Tables for Descriptive Statistics and Regression Models},
  author = {Amal Tawfik},
  year = {2026},
  note = {R package version 0.14.0},
  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.

mirror server hosted at Truenetwork, Russian Federation.