Skip to contents

Nice Spaghetti Plot for Two Within-Subject Conditions

Source: R/nice_spaghetti.R
nice_spaghetti.Rd

Creates a publication-ready paired (within-subject) spaghetti plot. Individual participant trajectories are shown as connected lines with optional mean overlay. Designed for pre-post designs and other two-condition repeated measures outcomes.

Usage

nice_spaghetti(
  data,
  pre,
  post,
  group = NULL,
  colors = NULL,
  pre_label = "Pre",
  post_label = "Post",
  y_label = NULL,
  title = NULL,
  subtitle = NULL,
  point_size = 2,
  line_alpha = 0.35,
  point_alpha = 0.5,
  jitter = 0,
  show_mean = FALSE
)

Arguments

data

A data frame in wide format containing the pre and post variables.

pre

Unquoted name of the pre-intervention variable.

post

Unquoted name of the post-intervention variable.

group

Optional character string specifying a grouping variable. When provided, individual trajectories and (if requested) mean lines are colored by group.

colors

Optional named character vector of colors for the group variable. Names must match group levels.

pre_label

Character. Label displayed for the pre condition.

post_label

Character. Label displayed for the post condition.

y_label

Character. Label for the y-axis. If NULL, a label is inferred from pre and post.

title

Character. Optional plot title.

subtitle

Character. Optional plot subtitle.

point_size

Numeric. Size of individual points. Default is 2.

line_alpha

Numeric between 0 and 1. Transparency of paired lines.

point_alpha

Numeric between 0 and 1. Transparency level for individual points. Lower values are recommended for large samples to reduce overplotting.

jitter

Numeric. Horizontal jitter width applied to individual points (no vertical jitter). Useful for large samples to improve visibility of overlap.

show_mean

Logical. If TRUE, overlays the mean trajectory.

Value

A ggplot object.

Details

The function reshapes wide data internally using tidy evaluation. It is intended for within-subject comparisons (e.g., pre-post RCT outcomes). No axis limits are imposed; scaling adapts to the data range.

Examples

if (requireNamespace("dplyr", quietly = TRUE) &&
    requireNamespace("tidyr", quietly = TRUE) &&
    requireNamespace("ggplot2", quietly = TRUE)) {

  df <- data.frame(
    pre_var = rnorm(50, 25, 5),
    post_var = rnorm(50, 15, 5)
  )

  nice_spaghetti(
    df,
    "pre_var",
    "post_var",
    pre_label = "Before",
    post_label = "After",
    title = "Reduction in Affective Polarization",
    subtitle = "Individual trajectories and group mean",
    show_mean = TRUE
  )
}