EFAtools (version 1.1.0)

print.efa: Print and summarise an efa object

Description

print() shows a concise overview of an efa_fit() or efa_mi() solution: a model header, the loading matrix (with the factor intercorrelations for oblique solutions), the variances accounted for, and the model fit. summary() returns a summary.efa object whose print method adds the full diagnostics: model and simple-structure diagnostics, confidence-interval tables, the structure matrix, multiple-imputation uncertainty (for pooled objects), and residual diagnostics. format() assembles the same report and returns it as a character vector; print() is cat(format(x), sep = "\n"). The lines follow the active console theme, so they are plain when colours are disabled (for example when captured into a file or stripped with cli::ansi_strip()).

Usage

# S3 method for efa
print(x, ...)
# S3 method for efa_mi
print(x, ...)
# S3 method for efa
format(
  x,
  cutoff = 0.3,
  digits = 3,
  max_name_length = 10,
  sort_loadings = c("none", "primary", "clustered"),
  show_loading_legend = TRUE,
  max_factors_per_block = NULL,
  ...
)
# S3 method for efa_mi
format(x, ...)
# S3 method for efa
summary(
  object,
  cutoff = 0.3,
  digits = 3,
  max_name_length = 10,
  ci = c("auto", "none", "separate"),
  ci_filter = c("salient", "all", "nonzero"),
  diagnostics_top_n = 10,
  residual_cutoff = 0.1,
  residual_top_n = 10,
  show_structure = TRUE,
  sort_loadings = c("none", "primary", "clustered"),
  show_loading_legend = TRUE,
  cross_loading_cutoff = cutoff,
  min_primary_gap = 0.2,
  min_salient_per_factor = 3,
  max_factors_per_block = NULL,
  show_mi_diagnostics = NULL,
  ...
)
# S3 method for efa_mi
summary(object, ...)
# S3 method for summary.efa
print(x, ...)
# S3 method for summary.efa
format(x, ...)

Value

print() and the print method for summary.efa objects return their argument invisibly. format() returns a character vector with the report lines. summary() returns an object of class summary.efa.

Arguments

x, object

An object of class efa (from efa_fit()) or efa_mi (from efa_mi()); for the summary.efa methods, the object returned by summary().

...

Further arguments passed to print.efa_loadings().

cutoff

numeric. The absolute value at or above which loadings are emphasised in the loading table. Default is .3.

digits

numeric. Number of decimal places for the printed tables. Default is 3.

max_name_length

numeric. Maximum length of the variable names to display; longer names are cut from the right, or abbreviated where cutting would give two variables the same label. Applies to every table that names variables. name_style (see print.efa_loadings()) can be passed through ... to choose the shortening explicitly, but it reaches the loading table only; the confidence-interval and simple-structure tables always shorten by cutting.

sort_loadings

character. Optional row sorting for the loading table. See print.efa_loadings().

show_loading_legend

logical. Whether to print a short legend for the loading-table styling. Default is TRUE. The legend is only printed when that styling is actually rendered (a colour-capable console); in plain output it is omitted and this argument has no effect.

max_factors_per_block

numeric or NULL. Maximum number of factor columns per loading-table block. If NULL, chosen from the console width.

ci

character. Which confidence intervals summary() shows, if available. "auto" and "separate" print CI sections when CIs were computed; "none" suppresses them. Default is "auto".

ci_filter

character. Which loading CIs summary() prints: "salient" (default), "all", or "nonzero"; see Details.

diagnostics_top_n

numeric. Maximum number of item-level entries summary() prints per simple-structure diagnostic.

residual_cutoff

numeric. Absolute residual cutoff for the residual diagnostics in summary(). Default is .1.

residual_top_n

numeric. Maximum number of residuals summary() prints. Use Inf to print all residuals above residual_cutoff.

show_structure

logical. Whether summary() prints the structure matrix for oblique solutions when available. Default is TRUE.

cross_loading_cutoff

numeric. Cutoff for counting cross-loadings in the summary() diagnostics. Defaults to cutoff.

min_primary_gap

numeric. Minimum desired absolute difference between the largest and second-largest absolute loading of an item, used in the summary() diagnostics.

min_salient_per_factor

numeric. Minimum number of salient indicators per factor used in the summary() diagnostics. Default is 3.

show_mi_diagnostics

logical or NULL. Whether summary() prints a multiple-imputation uncertainty summary for pooled EFAs. NULL shows it for pooled objects.

Details

The methods are shared by single-imputation efa objects and pooled efa_mi objects. For efa_mi objects the header reports the number of imputations and the alignment/pooling settings; confidence intervals and a multiple-imputation uncertainty summary are shown by summary() when the pooled object carries bootstrap/MI quantities.

In summary(), ci_filter controls which loading intervals are shown: "salient" reports intervals for loadings whose absolute point estimate is at least cutoff, "nonzero" reports intervals excluding zero, and "all" reports every finite interval.

Examples

mod <- efa_fit(test_models$baseline$cormat, n_factors = 3, N = 500,
               estimator = "PAF", rotation = "promax")
mod

# The full diagnostics, CI tables, and residual diagnostics:
summary(mod)

# format() returns the same lines as a character vector, e.g. for a report file:
writeLines(format(mod))