Tidying methods for a spicy_continuous_lm_table

Standard broom::tidy() and broom::glance() interfaces for an object returned by table_continuous_lm(). They re-shape the underlying long-format data into the two canonical broom views so the table can be consumed by gtsummary, modelsummary, parameters, and any other tidyverse-stats pipeline.

Usage

# S3 method for class 'spicy_continuous_lm_table'
tidy(x, ...)

# S3 method for class 'spicy_continuous_lm_table'
glance(x, ...)

Arguments

x

A spicy_continuous_lm_table returned by table_continuous_lm().

...

Currently ignored. Present for compatibility with the broom::tidy() / broom::glance() generics.

Value

A tbl_df (when tibble is installed) or a plain data.frame.

Details

tidy() returns one row per estimated parameter across all outcomes:

• One row per fitted level mean (estimate_type = "emmean") for categorical predictors, with the level name in term.

• One row per contrast (estimate_type = "difference") when a binary contrast is shown, with term = "<level2> - <level1>".

• One row per slope (estimate_type = "slope") for numeric predictors, with term = predictor_label.

Standard broom columns: outcome, label, term, estimate_type, estimate, std.error, conf.low, conf.high, statistic, p.value. The outcome column carries the original variable name; label carries the human-readable label.

glance() returns one row per outcome with model-level statistics. Columns: outcome, label, predictor_type ("categorical" or "continuous"), test_type ("F" for categorical predictors, "t" for continuous ones), statistic, df, df.residual, p.value, r.squared, adj.r.squared, es_type, es_value, es_ci_lower, es_ci_upper, nobs, weighted_n.

See also

as.data.frame.spicy_continuous_lm_table() for the raw long-format access.