Print an efa_sl_loadings object
# S3 method for efa_sl_loadings
print(x, ...)# S3 method for efa_sl_loadings
format(
x,
cutoff = 0.2,
digits = 3,
max_name_length = 10,
color = TRUE,
name_style = c("truncate", "abbreviate", "full"),
max_factors_per_block = NULL,
sort_loadings = c("none", "primary", "clustered"),
...
)
print() returns its argument x invisibly; it is
cat(format(x, ...), sep = "\n") followed by a blank line for console
spacing. format() returns a character vector with the table lines (styled
to the active console theme; plain when colours are disabled).
class efa_sl_loadings matrix.
additional arguments passed to print or format.
numeric. The value at or above which loadings are emphasized
(default is .2). The default is lower than the .3 of an ordinary loading
table (print.efa_loadings()): the group-factor loadings are residualized,
that is, they carry only the variance left once the general factor has been
partialled out, and are therefore smaller than the corresponding first-order
loadings.
numeric. Passed to round. Number of digits
to round the loadings to (default is 3).
numeric. The maximum length of the variable names to
display; see print.efa_loadings().
logical. Whether to apply console styling using cli.
Default is TRUE.
character. How to shorten variable names longer than
max_name_length; see print.efa_loadings().
numeric or NULL. Maximum number of factor
columns to print per block. If NULL, the number is chosen from the
console width.
character. Optional row sorting; see
print.efa_loadings(). The default "none" keeps the input order. When
sorting is requested, rows are grouped by their largest group-factor
loading: the general factor is left out of the comparison, since it is the
largest loading of almost every item and sorting on it would leave the order
untouched.
Prints a Schmid-Leiman loading matrix (general factor, group factors, and the
communality/uniqueness columns) as a styled, decimal-aligned table. Loadings with
absolute value greater than or equal to cutoff are emphasised, smaller loadings are
de-emphasised, and Heywood-relevant cells (a loading or communality above 1, or a
negative uniqueness) are highlighted. If the matrix has many columns or the console is
narrow, the table is split into stacked column blocks so the output stays readable.
EFA_mod <- efa_fit(test_models$baseline$cormat, N = 500, n_factors = 3,
estimator = "PAF", rotation = "promax")
efa_schmid_leiman(EFA_mod, estimator = "PAF")
Run the code above in your browser using DataLab