Why inspect variables before analysis? | Recover labels from imported column names | Inspect variables with varlist() | Select subsets of variables | Build an interactive codebook | When to use varlist() and code_book()

Frequency tables with freq() | Basic usage | Sorting | Cumulative percentages | Weighted frequencies | Labelled variables | Custom missing values | Declared missing values | Cross-tabulations with cross_tab() | Basic two-way table | Row and column percentages | Grouping with by | Ordinal variables | Confidence intervals for effect sizes | Weighted cross-tabulations | Monte Carlo simulation | Data frame output | Setting global defaults | Learn more

Choosing the right measure | Quick overview with assoc_measures() | Nominal variables | Cramer's V | Phi coefficient | Contingency coefficient | Ordinal variables | Goodman-Kruskal Gamma | Kendall's Tau-b | Kendall's Tau-c | Somers' D | Asymmetric (PRE) measures | Lambda | Goodman-Kruskal Tau | Uncertainty coefficient | Yule's Q | Automatic selection in cross_tab() | Confidence intervals | Controlling decimal places | See also | References

Basic usage | One-way tables | Output formats | gt output | tinytable output | Data frame output | Custom labels | Association measures and confidence intervals | Weighted tables | Handling missing values | Filtering and reordering levels | Formatting options | Decimal alignment | Tidying for downstream pipelines | Exporting to Excel, Word, or clipboard | See also

Choose the right function | A shared interface | A practical reporting sequence | Categorical variables | Continuous variables | Model-based continuous variables | The coefficient table | Choose the output format | Post-process the returned table object | Keep the detailed options in the function-specific articles

Inspect your data | Frequency tables | Cross-tabulations | Association measures | Summary tables | Row-wise summaries | Learn more

Basic usage | Robust standard errors | Cluster-robust standard errors | Bootstrap and jackknife | Case weights | Numeric predictors | Covariate adjustment | Two estimands for the adjusted marginal means | Effect sizes under adjustment | Scope: additive only in v1 | Effect sizes | Confidence intervals for effect sizes | Article-style polish | Publication-ready output | Tidying for downstream pipelines | See also | References

Basic usage | Grouped summaries | Add test statistics and effect sizes | Selecting variables | Handling missing values | Labels and output formats | Export to Excel or Word | Display options | Tidying for downstream pipelines | See also

Basic usage | Standardised coefficients | Per-coefficient effect sizes | Average marginal effects (AME) | Multiple models side by side | Univariable screening | Hierarchical / nested regression | Robust variance | Heteroskedasticity-consistent (HC) | Cluster-robust (CR) | Multiple-comparison adjustment | Filtering displayed coefficients | Significance stars | Display options | Generalised linear models (glm) | Response-scale display: exponentiate = TRUE | Event counts per level | Term-level partial chi-square | Standardised coefficients: refit and pseudo | Average marginal effects: probability units, not log-odds | Profile-likelihood CIs: ci_method = "profile" | Hierarchical glm (LRT) | Gaussian glm caveat | Mixed-effects models | Output formats | broom integration | See also | References

What changes in a Bayesian table | Check convergence before reading anything | Priors are part of the model — disclose them | Odds ratios: exponentiate = TRUE | Marginal effects, from the draws | The frequentist twin, side by side | Fit statistics for a posterior | Standardized coefficients from the draws | What is refused, and why | Multilevel models: random effects from the draws | brms | Output formats | See also | References

Count models in one paragraph | Poisson baseline | Overdispersion: the negative binomial | Excess zeros: zero-inflation and hurdle | Which count model? Side by side | One number per predictor: the combined AME | Cluster-robust variance | Mixed counts: glmmTMB | Counts with high-dimensional fixed effects: fixest | Output formats | References

The mixed model in one paragraph | Basic table | Why the variance components carry no p-value | From OLS to a multilevel model, step by step | Step 0: the regression you would have fitted anyway | Step 1: the empty model — how much do schools matter? | Step 2: a random intercept, next to the OLS | Step 3: one variable, two effects — the contextual split | Step 4: does the SES slope vary across schools? | Step 5: why do slopes differ? A cross-level interaction | Three levels, and crossed designs | When the fit is singular | Testing individual components: re_test | The random-effects block on your terms | Random intercept or random slope? Models side by side | Generalized linear mixed models | Average marginal effects | glmmTMB: beyond lme4 | nlme::lme | Cluster-robust and other variance estimators | Standardized coefficients | Output formats | Where next | References

The baseline-category logit model in one paragraph | Basic table | Odds ratios: exponentiate = TRUE | Changing the reference category | Does a predictor matter at all? | Can two categories be combined? | The sign trap, and marginal effects | An ordinal predictor: scores or dummies? | Independence of irrelevant alternatives, briefly | Discrete choice: mlogit | Several models side by side | Output formats | References

The proportional-odds model | Basic table | Odds ratios: exponentiate = TRUE | Average marginal effects: a probability matrix | Cluster-robust standard errors | Standard errors and confidence intervals | The ordinal::clm() engine | Testing proportional odds | Partial proportional odds: nominal = ~ | Scale effects: scale = ~ | Choosing an ordinal specification | Several models side by side | Output formats | References

The registry | Cross-cutting arguments | The families in brief | When a class is not supported | Programmatic access

The Cox model in one paragraph | Basic table: hazard ratios | Why there is no AME column | Absolute effects: RMST and risk differences | Multi-centre data: cluster-robust variance | Hierarchical Cox models | The univariable screen | Accelerated failure time: survreg() | rms::cph() | Fully parametric: flexsurv | Output formats | References

Parse | Audit | Render | Useful arguments | PDF output | Audit-only document

How a factor enters the model | Choosing and changing the reference level | Does the factor matter at all? The joint test | Ordered factors: R's polynomial contrasts | Scores or dummies? | A third coding: successive differences | The reverse temptation: categorizing a continuous predictor | Presentation | See also | References