surveywts provides tidy tools for calibrating survey weights to known population totals, adjusting for nonresponse, and diagnosing weight quality — all with full weighting history tracking for reproducible survey analysis.
Installation
# From GitHub (development version)
pak::pak("JDenn0514/surveywts")
# From r-universe (pre-built binaries, no GitHub PAT needed)
install.packages("surveywts", repos = "https://jdenn0514.r-universe.dev")Overview
surveywts is part of the surveyverse ecosystem. It provides three calibration methods, nonresponse adjustment, and weight diagnostics — all using tidy, formula-free syntax.
| Function | Purpose |
|---|---|
calibrate() |
GREG calibration to population totals |
rake() |
Raking (iterative proportional fitting) |
poststratify() |
Post-stratification to cell counts or proportions |
adjust_nonresponse() |
Nonresponse adjustment via weighting classes |
effective_sample_size() |
ESS (Kish approximation) |
weight_variability() |
CV and design effect of weights |
summarize_weights() |
Summary statistics, optionally by group |
Every function tracks the full weighting history so you can audit exactly what transformations were applied and in what order.
Learn more
Full documentation is available at https://jdenn0514.github.io/surveywts/.