Plot a cross-level fit

"profiles" shows the individual profiles' means per indicator; "group_means" the group classes' means of the group means; "composition" the profile shares within each group class. Profiles and classes are told apart by colour and point shape.

Usage

# S3 method for class 'multilpa_cross_level'
plot(
  x,
  what = c("profiles", "group_means", "composition"),
  main = NULL,
  subtitle = NULL,
  ...
)

Arguments

x

A multilpa_cross_level fit.

what

Which view.

main, subtitle

Optional title and subtitle.

...

Unused.

Value

A ggplot object. Raises latents_missing_package without ggplot2.

Examples

set.seed(1)
ratings <- data.frame(
  team = rep(seq_len(40), each = 6),
  climate = rep(rnorm(40, rep(c(-1, 1), each = 20), 0.5), each = 6) +
    rnorm(240, sample(c(-1, 1), 240, replace = TRUE), 0.6))
fit <- multilpa(ratings, "climate", "team", n_profiles = 2,
                n_group_classes = 2, family = "full_cross_level",
                n_starts = 3, seed = 1)
if (requireNamespace("ggplot2", quietly = TRUE)) plot(fit)
#> `geom_line()`: Each group consists of only one observation.
#> ℹ Do you need to adjust the group aesthetic?