plot_models_fm {semhelpinghands}

Plot Models on a Chi-Squares-vs-Dfs Graph

Description

Plot models on a graph with model chi-square against model the degrees of freedom, with lines for equal fit measures.

Usage

plot_models_fm(
  ...,
  fit_measure = c("cfi", "tli", "rmsea"),
  fit_values,
  line_size = 1,
  label_size = 8,
  point_size = 5,
  position_dodge = 0.5,
  include_model_values = FALSE,
  include_baseline = FALSE
)

Arguments

...

The lavaan::lavaan objects to be plotted. Can also be a named list of the lavaan::lavaan objects. If it is as list, it must be named and the names will be used in the plot.

fit_measure

A length-one character vector of the fit measures to use to plot the lines. Only supports "cfi" (the default), "tli", and "rmsea".

fit_values

A numeric vector of the values of the fit measure used to plot the lines. The default values are c(.90, .95) for "cfi" and "tli", and c(.00, .02, .05, .08) for "rmsea".

line_size

The size of the lines. Default is 1.

label_size

The size of the model names. Default is 8.

point_size

The size of the point representing a model. Default is 2.

position_dodge

Offsetting the label of a model from the point. Default is .5. Used by ggrepel::geom_label_repel().

include_model_values

If TRUE , the values of the models on fit_measure will be added to fit_values. Default is FALSE.

include_baseline

If TRUE, the baseline model is included in the plot. Default is FALSE.

Details

This function plots models based on their model chi-squares and model degrees of freedoms.It can also add lines for chi-square-df combination with equal values on selected fit measures. Currently supports CFI, TLI, and RMSEA.

Value

Return a ggplot2::ggplot() output that can be further modified.

Author(s)

Shu Fai Cheung https://orcid.org/0000-0002-9871-9448

See Also

lavaan::fitMeasures()

Examples

library(lavaan)

# From the help page of modificationIndices

HS.model <- '
 visual  =~ x1 + x2 + x3
 textual =~ x4 + x5 + x6
 speed   =~ x7 + x8 + x9
'

fit <- cfa(HS.model, data = HolzingerSwineford1939)
modindices(fit, sort = TRUE, op = "=~")

fit2 <- update(fit, add = "visual =~ x9")
fit3 <- update(fit, add = "textual =~ x3\nvisual =~ x7")

models <- list(Initial = fit,
               Model_2 = fit2,
               Model_3 = fit3)
fit_cfi <- sapply(models, fitMeasures, fit.measures = "cfi")
fit_tli <- sapply(models, fitMeasures, fit.measures = "tli")
fit_rmsea <- sapply(models, fitMeasures, fit.measures = "rmsea")

# Supply the models as arguments
plot_models_fm(fit, fit2, fit3)

# Plot lines for selected values on a fit measure (CFI by default)
plot_models_fm(fit, fit2, fit3, fit_values = c(.90, .925, .95, fit_cfi))

# Plot the models' values on the fit measures
plot_models_fm(fit, fit2, fit3, include_model_values = TRUE)

# Supply the models as a named list
plot_models_fm(list(A = fit, B = fit2, C = fit3),
               fit_values = c(.90, .925, .95))

# Plot the models, fit measure set to TLI
plot_models_fm(fit, fit2, fit3, fit_measure = "tli")
plot_models_fm(fit, fit2, fit3, fit_measure = "tli",
               fit_values = c(.90, .925, .95, fit_tli))
plot_models_fm(fit, fit2, fit3, fit_measure = "tli",
               include_model_values = TRUE)

# Plot the models, fit measure set to RMSEA
plot_models_fm(fit, fit2, fit3, fit_measure = "rmsea")
plot_models_fm(fit, fit2, fit3, fit_measure = "rmsea",
               include_model_values = TRUE)