safe_resid_position: Adjust Residual Positions

Description

Rotate the residuals (or R-squares) to avoid overlapping with paths.

Usage

safe_resid_position(
  object,
  layout,
  default_angle = 0,
  style = c("1200", "geometry"),
  update_plot = TRUE
)

Value

If object is a lavaan-class object, it returns a named numeric vector of residual angles to be used by rotate_resid(). If object is a qgraph object and update_plot is TRUE, it returns a qgraph object with the residuals's angles adjusted. Otherwise, it returns a named vector of the angles, as for a lavaan-class object.

Arguments

object: It can be the output of lavaan::sem() or lavaan::lavaan(), or a lavaan-class object. The model must have a beta matrix of the structural path. It can also be a qgraph object generated by semPlot::semPaths().
layout: A layout matrix. Required if object is a lavaan-class object. Ignored if object is a qgraph object.
default_angle: Used if object is a lavaan-class object. The default position of a residual, defined in the same way angle is defined for rotate_resid(). Ignored if object is a qgraph object.
style: The convention for the angles. If "1200", the default, the convention of rotate_resid() is used: top (12 o'clock) is 0, clockwise angle is positive and counterclockwise angle is negative. if "geometry", then the convention in geometry is used: right is 0, counterclockwise is positive, and clockwise is negative.
update_plot: Logical. Used on if object is a qgraph object. If TRUE, the function returns a modified qgraph object. If FALSE, the function returns a named vector of the new positions.

Details

This function identify all directed paths connected to a node, and find the largest arc with no directed paths. The residual (or R-square) is then set to the mid-point of this arc.

This function is intended for having a "likely" readable graph with as little user-intervention as possible. If precise control of the positions is desired, use rotate_resid().

Only directed paths (single-headed arrows) will be considered. Bidirectional paths such as covariances are not taken into account.

Examples

Run this code


library(lavaan)
library(semPlot)
# Create a dummy dataset
mod_pa <-
"
m11 ~ x1
m21 ~ m11
m2 ~ m11
m22 ~ m11
y ~ m2 + m21 + m22 + x1
"
fit <- lavaan::sem(
          mod_pa,
          do.fit = FALSE
        )
dat <- simulateData(
          parameterTable(fit),
          sample.nobs = 500,
          seed = 1234
        )
fit <- lavaan::sem(
          mod_pa,
          dat
        )
# Set the layout
m <- auto_layout_mediation(
        fit
      )
p <- semPaths(
          fit,
          whatLabels = "est",
          layout = m,
          DoNotPlot = TRUE
        ) |>
      safe_edge_label_position()
plot(p)
# Update the plot
p_safe <- p |> safe_resid_position()
plot(p_safe)
# Set the positon manually
pos_new <- safe_resid_position(p,
                               update_plot = FALSE)
pos_new
p_safe2 <- p |>
            rotate_resid(pos_new)
plot(p_safe2)