Figures in R

1 Picking a Chart Type

https://www.data-to-viz.com

2 Gallery

https://www.r-graph-gallery.com

3 Resources for Learning `R` Syntax for Figures

https://www.statmethods.net/graphs/index.html

http://www.cookbook-r.com/Graphs/

4 Petersen Lab Examples

https://research-git.uiowa.edu/PetersenLab/R-Plotting/-/tree/main/Analyses

5 Font

You can download the lab fonts for figures here¹: https://drive.google.com/drive/u/0/folders/1fqlrnEe7NFnWZoIrsHmr8ulDS4nhs-H3

5.1 Preamble

5.1.1 Install Libraries

Code

#install.packages("remotes")
#remotes::install_github("DevPsyLab/petersenlab")

5.1.2 Load Libraries

Code

library("petersenlab")
library("ellipse")
library("ggplot2")
library("grid")
library("reshape")
library("plyr")
library("RColorBrewer")
library("reshape2")
library("ggExtra")
library("viridis")
library("ggthemes")
library("ggpubr")

5.2 Simulate Data

Code

set.seed(52242)

n <- 1000

predictor <- rbeta(n, 1.5, 5) * 100
outcome <- predictor + rnorm(n, mean = 0, sd = 20) + 50
number <- sample(1:1000, replace = TRUE)

predictorOverplot <- sample(1:50, n, replace = TRUE)
outcomeOverplot <- predictorOverplot + sample(1:75, n, replace = TRUE)

df <- data.frame(predictor = predictor,
                 outcome = outcome,
                 predictorOverplot = predictorOverplot,
                 outcomeOverplot = outcomeOverplot)

df[sample(1:n, size = 10), "predictor"] <- NA
df[sample(1:n, size = 10), "outcome"] <- NA
df[sample(1:n, size = 10), "predictorOverplot"] <- NA
df[sample(1:n, size = 10), "outcomeOverplot"] <- NA

5.3 Line

Code

plot.new()
lines(
  x = seq(from = -10, to = 10, length.out = 100),
  y = seq(from = -25, to = 25, length.out = 100))

5.4 Curve

Code

curve(x^3 - 3*x, from = -2, to = 2)
curve(x^2 - 2, add = TRUE, col = "violet")

5.5 Basic Scatterplot

5.5.1 Base R

Code

plot(outcome ~ predictor, data = df)

Code

plot(df$predictor, df$outcome)

5.5.1.1 Best-fit line

Code

plot(outcome ~ predictor, data = df)
abline(lm(outcome ~ predictor, data = df), col = "red") #regression line (y~x)

5.5.1.2 Best-fit line with correlation coefficient

Code

plot(outcome ~ predictor, data = df)
abline(lm(outcome ~ predictor, data = df), col = "red") #regression line (y~x)
addText(x = df$predictor, y = df$outcome)

5.5.1.3 Loess line

Code

plot(outcome ~ predictor, data = df)
lines(loess.smooth(df$predictor, df$outcome)) #loess line (x,y)

5.5.2 `ggplot2`

Code

ggplot(df, aes(x = predictor, y = outcome)) +
  geom_point() +
  theme_classic()

5.5.2.1 Best-fit line

Code

ggplot(df, aes(x = predictor, y = outcome)) +
  geom_point() +
  stat_smooth(method = "lm", formula = y ~ x) +
  theme_classic()

5.5.2.2 Best-fit line with correlation coefficient

Code

ggplot(df, aes(x = predictor, y = outcome)) +
  geom_point() +
  stat_smooth(method = "lm", formula = y ~ x) +
  stat_cor(
    cor.coef.name = "r",
    p.accuracy = 0.001,
    r.accuracy = 0.01) +
  theme_classic()

5.5.2.3 Loess line

Code

ggplot(df, aes(x = predictor, y = outcome)) +
  geom_point() +
  stat_smooth(method = "loess", formula = y ~ x) +
  theme_classic()

5.6 Change Plot Style

5.6.1 Change Theme

Code

basePlot <- ggplot(df, aes(x = predictor, y = outcome)) +
  geom_point()

5.6.1.1 Default Theme

Code

basePlot

5.6.1.2 Grayscale: `theme_gray()`

Code

basePlot + theme_gray() + theme(text = element_text(family = "Gotham"))

5.6.1.3 Black-and-White: `theme_bw()`

Code

basePlot + theme_bw() + theme(text = element_text(family = "Gotham"))

5.6.1.4 Line Drawing: `theme_linedraw()`

A theme with only black lines of various widths on white backgrounds, reminiscent of a line drawing. Note that this theme has some very thin lines (<< 1 pt) which some journals may refuse.

Code

basePlot + theme_linedraw() + theme(text = element_text(family = "Gotham"))

5.6.1.5 Light: `theme_light()`

Code

basePlot + theme_light() + theme(text = element_text(family = "Gotham"))

5.6.1.6 Dark: `theme_dark()`

Code

basePlot + theme_dark() + theme(text = element_text(family = "Gotham"))

5.6.1.7 Minimal: `theme_minimal()`

Code

basePlot + theme_minimal() + theme(text = element_text(family = "Gotham"))

5.6.1.8 Classic: `theme_classic()`

Code

basePlot + theme_classic() + theme(text = element_text(family = "Gotham"))

5.6.1.9 A Completely Empty Theme: `theme_void()`

Code

basePlot + theme_void() + theme(text = element_text(family = "Gotham"))

5.6.1.10 Visual Unit Tests: `theme_test()`

Code

basePlot + theme_test() + theme(text = element_text(family = "Gotham"))

5.6.1.11 Edward Tufte: `theme_tufte()`

Theme based on Edward Tufte.

Code

basePlot + theme_tufte()

5.6.1.12 Wall Street Journal: `theme_wsj()`

Theme based on the publication, the Wall Street Journal.

Code

basePlot + theme_wsj()

5.6.1.13 FiveThirtyEight: `theme_fivethirtyeight()`

Theme based on the publication, FiveThirtyEight.

Code

basePlot + theme_fivethirtyeight()

5.6.1.14 The Economist: `theme_economist()`

Theme based on the publication, The Economist.

Code

basePlot + theme_economist()

5.6.1.15 Stephen Few: `theme_few()`

Theme based on the rules and examples from Stephen Few’s Show Me the Numbers and “Practical Rules for Using Color in Charts”.

Code

basePlot + theme_few()

5.7 Add Marginal Distributions

Code

scatterplot <- ggplot(df, aes(x = predictor, y = outcome)) +
  geom_point() +
  theme_classic() +
  theme(text = element_text(family = "Gotham"))

5.7.1 Density Plot

Code

densityMarginal <- ggMarginal(scatterplot, type = "density", xparams = list(fill = "gray"), yparams = list(fill = "gray"))

Code

print(densityMarginal, newpage = TRUE)

5.7.2 Histogram

Code

histogramMarginal <- ggMarginal(scatterplot, type = "histogram", xparams = list(fill = "gray"), yparams = list(fill = "gray"))

Code

print(histogramMarginal, newpage = TRUE)

5.7.3 Boxplot

Code

boxplotMarginal <- ggMarginal(scatterplot, type = "boxplot", xparams = list(fill = "gray"), yparams = list(fill = "gray"))

Code

print(boxplotMarginal, newpage = TRUE)

5.7.4 Violin Plot

Code

violinMarginal <- ggMarginal(scatterplot, type = "violin", xparams = list(fill = "gray"), yparams = list(fill = "gray"))

Code

print(violinMarginal, newpage = TRUE)

5.7.5 Density Plot and Histogram

Code

densigramMarginal <- ggMarginal(scatterplot, type = "densigram", xparams = list(fill = "gray"), yparams = list(fill = "gray"))

Code

print(densigramMarginal, newpage = TRUE)

5.8 Ellipse

5.8.1 Basic Ellipse

Code

ggplot(df, aes(x = predictor, y = outcome)) +
  geom_point() +
  stat_ellipse(alpha = 0.4, level = 0.95, geom = "polygon", fill = "red", color = "red") +
  theme_classic() +
  theme(text = element_text(family = "Gotham"))

5.8.2 Align Coordinates

Code

ggplot(df, aes(x = predictor, y = outcome)) +
  geom_point() +
  stat_ellipse(alpha = 0.4, level = 0.95, geom = "polygon", fill = "red", color = "red") +
  scale_x_continuous(expand = c(0, 0)) +
  scale_y_continuous(expand = c(0, 0)) +
  coord_fixed(ratio = (max(predictor, na.rm = TRUE) - min(predictor, na.rm = TRUE))/(max(outcome, na.rm = TRUE) - min(outcome, na.rm = TRUE)),
              xlim = c(0, max(predictor, na.rm = TRUE)), 
              ylim = c(0, max(outcome, na.rm = TRUE))) +
  theme_classic() +
  theme(text = element_text(family = "Gotham"))

5.8.3 Reduce Dot Size

Code

ggplot(df, aes(x = predictor, y = outcome)) +
  geom_point(size = 0.5) +
  stat_ellipse(alpha = 0.4, level = 0.95, geom = "polygon", fill = "red", color = "red") +
  scale_x_continuous(expand = c(0, 0)) +
  scale_y_continuous(expand = c(0, 0)) +
  coord_fixed(ratio = (max(predictor, na.rm = TRUE) - min(predictor, na.rm = TRUE))/(max(outcome, na.rm = TRUE) - min(outcome, na.rm = TRUE)),
              xlim = c(0, max(predictor, na.rm = TRUE)), 
              ylim = c(0, max(outcome, na.rm = TRUE))) +
  theme_classic() +
  theme(text = element_text(family = "Gotham"))

5.8.4 Transparency

Code

ggplot(df, aes(x = predictor, y = outcome)) +
  geom_point(alpha = 0.3) +
  stat_ellipse(alpha = 0.4, level = 0.95, geom = "polygon", fill = "red", color = "red") +
  scale_x_continuous(expand = c(0, 0)) +
  scale_y_continuous(expand = c(0, 0)) +
  coord_fixed(ratio = (max(predictor, na.rm = TRUE) - min(predictor, na.rm = TRUE))/(max(outcome, na.rm = TRUE) - min(outcome, na.rm = TRUE)),
              xlim = c(0, max(predictor, na.rm = TRUE)), 
              ylim = c(0, max(outcome, na.rm = TRUE))) +
  theme_classic() +
  theme(text = element_text(family = "Gotham"))

5.9 Bubble Chart

5.9.1 Basic Bubble Chart

Code

ggplot(df, aes(x = predictorOverplot, y = outcomeOverplot)) +
  geom_count(aes(size = ..n..)) +
  scale_size_area() +
  theme_classic() +
  theme(text = element_text(family = "Gotham"))

5.9.2 Specify Sizes

Code

ggplot(df, aes(x = predictorOverplot, y = outcomeOverplot)) +
  geom_count(aes(size = ..n..)) +
  scale_size_continuous(breaks = c(1, 2, 3, 4), range = c(1, 7)) +
  theme_classic() +
  theme(text = element_text(family = "Gotham"))

5.10 2-Dimensional Density

Code

ggplot(df, aes(x = predictor, y = outcome)) +
  stat_density_2d(aes(fill = ..density..), geom = "raster", contour = FALSE) +
  scale_x_continuous(expand = c(0, 0)) +
  scale_y_continuous(expand = c(0, 0)) +
  scale_fill_viridis() +
  theme(
    legend.position = "none",
    text = element_text(family = "Gotham")
  )

5.11 Combined Ellipse and Bubble Chart

5.11.1 `ggplot2`

Code

ggplot(df, aes(x = predictorOverplot, y = outcomeOverplot)) +
  geom_count(alpha = .6, color = rgb(0,0,.7,.5)) +
  scale_size_continuous(breaks = c(1, 2, 3, 4), range = c(1, 7)) +
  stat_smooth(method = "loess", se = TRUE, color = "green") + 
  stat_smooth(method = "lm") +
  stat_ellipse(alpha = 0.4, level = 0.95, geom = "polygon", fill = "red", color = "red") +
  scale_x_continuous(expand = c(0, 0)) +
  scale_y_continuous(expand = c(0, 0)) +
  coord_fixed(ratio = (max(predictorOverplot, na.rm = TRUE) - min(predictorOverplot, na.rm = TRUE))/(max(outcomeOverplot, na.rm = TRUE) - min(outcomeOverplot, na.rm = TRUE)),
              xlim = c(0, max(predictorOverplot, na.rm = TRUE)), 
              ylim = c(0, max(outcomeOverplot, na.rm = TRUE))) +
  theme_classic() +
  theme(text = element_text(family = "Gotham"))

Code

ggplot(df, aes(x = predictorOverplot, y = outcomeOverplot)) +
  geom_count(alpha = .6, color = rgb(0,0,.7,.5)) +
  scale_size_continuous(breaks = c(1, 2, 3, 4), range = c(1, 7)) +
  stat_smooth(method = "loess", se = TRUE, color = "green") + 
  stat_smooth(method = "lm") +
  stat_ellipse(color = "red", size = 1.5) +
  scale_x_continuous(expand = c(0, 0)) +
  scale_y_continuous(expand = c(0, 0)) +
  coord_fixed(ratio = (max(predictorOverplot, na.rm = TRUE) - min(predictorOverplot, na.rm = TRUE))/(max(outcomeOverplot, na.rm = TRUE) - min(outcomeOverplot, na.rm = TRUE)),
              xlim = c(0, max(predictorOverplot, na.rm = TRUE)), 
              ylim = c(0, max(outcomeOverplot, na.rm = TRUE))) +
  theme_classic() +
  theme(text = element_text(family = "Gotham"))

5.11.2 Other implementation

From: https://stats.stackexchange.com/questions/7899/complex-regression-plot-in-r

5.11.2.1 `ggplot2`

Code

df$x <- df$predictorOverplot
df$y <- df$outcomeOverplot

xc <- with(df, xyTable(x, y))
df2 <- cbind.data.frame(x = xc$x, y = xc$y, number = xc$number)
df2$n <- cut(df2$number, c(0,1.5,2.5,Inf), labels = c(1,2,4))
df.ell <- as.data.frame(with(df, ellipse(cor(df$x, df$y, use = "pairwise.complete.obs"),
                                         scale = c(sd(df$x, na.rm = TRUE), sd(df$y, na.rm = TRUE)),
                                         centre = c(mean(df$x, na.rm = TRUE), mean(df$y, na.rm = TRUE)),
                                         level = .95)))

ggplot(data = na.omit(df2), aes(x = x, y = y)) + 
  geom_point(aes(size = n), alpha = .6, color = rgb(0,0,.7,.5)) + 
  stat_smooth(data = df, method = "loess", se = FALSE, color = "green") + 
  stat_smooth(data = df, method = "lm", col = "red") +
  geom_path(data = df.ell, colour = "green", size = 1) +
  coord_cartesian(xlim = c(-1,60), ylim = c(-1,130))

5.11.2.2 Base `R`

Code

do.it <- function(df, type="confidence", ...) {
  require(ellipse)
  lm0 <- lm(y ~ x, data=df)
  xc <- with(df, xyTable(x, y))
  df.new <- data.frame(x = seq(min(df$x), max(df$x), 0.1))
  pred.ulb <- predict(lm0, df.new, interval = type)
  pred.lo <- predict(loess(y ~ x, data = df), df.new)
  plot(xc$x, xc$y, cex = xc$number*1/4, xlab = "x", ylab = "y", ...) #change number*X to change dot size
  abline(lm0, col = "red")
  lines(df.new$x, pred.lo, col="green", lwd = 2)
  lines(df.new$x, pred.ulb[,"lwr"], lty = 2, col = "red")
  lines(df.new$x, pred.ulb[,"upr"], lty = 2, col = "red")    
  lines(ellipse(cor(df$x, df$y), scale=c(sd(df$x),sd(df$y)), 
                centre = c(mean(df$x), mean(df$y)), level = .95), lwd = 2, col = "green")
  invisible(lm0)
}

df3 <- na.omit(df[sample(nrow(df), nrow(df), rep = TRUE),])
df3$x <- df3$predictorOverplot
df3$y <- df3$outcomeOverplot

do.it(df3, pch = 19, col = rgb(0,0,.7,.5))

5.12 Visually-Weighted Regression

5.12.1 Default

Code

vwReg(outcome ~ predictor, data = df)

5.12.2 Shade

Code

vwReg(outcome ~ predictor, data = df, shade = TRUE, spag = FALSE, show.lm = TRUE, show.CI = TRUE, bw = FALSE, B = 1000, quantize = "continuous")

Code

vwReg(outcome ~ predictor, data = df, shade = TRUE, spag = FALSE, show.lm = TRUE, show.CI = TRUE, bw = FALSE, B = 1000, quantize = "SD")

5.12.3 Spaghetti

Code

vwReg(outcome ~ predictor, data = df, shade = FALSE, spag = TRUE, show.lm = TRUE, show.CI = TRUE, bw = FALSE, B = 1000)

Code

vwReg(outcome ~ predictor, data = df, shade = FALSE, spag = TRUE, show.lm = FALSE, show.CI = FALSE, bw = FALSE, B = 1000)

5.12.4 Black/white

Code

vwReg(outcome ~ predictor, data = df, shade = TRUE, spag = FALSE, show.lm = TRUE, show.CI = TRUE, bw = TRUE, B = 1000, quantize = "continuous")

Code

vwReg(outcome ~ predictor, data = df, shade = TRUE, spag = FALSE, show.lm = TRUE, show.CI = TRUE, bw = TRUE, B = 1000, quantize = "SD")

Code

vwReg(outcome ~ predictor, data = df, shade = FALSE, spag = TRUE, show.lm = TRUE, show.CI = TRUE, bw = TRUE, B = 1000, quantize = "SD")

6 Graphic Design Principles for Data Visualization

https://www.data-to-viz.com/caveats.html

7 Types of Plots

7.1 Univariate Distribution

Used for: distribution of one numeric variable

7.1.1 Gallery

Violin chart: https://r-graph-gallery.com/violin.html
Density chart: https://r-graph-gallery.com/density-plot.html
Histogram: https://r-graph-gallery.com/histogram.html
Boxplot: https://r-graph-gallery.com/boxplot.html
Ridgeline chart: https://r-graph-gallery.com/ridgeline-plot.html

7.2 Bivariate Scatterplots

Used for: association between two numeric variables

7.2.1 Base `R`

plot(x, y)

7.2.2 ggplot2 package

http://www.cookbook-r.com/Graphs/Scatterplots_(ggplot2)/

ggplot(data, aes(x, y)) +
    geom_point()

7.3 Bivariate Barplots

Used for: association between one categorical variable and one numeric variable (or for depicting the frequency of categories of a categorical variable)

7.3.1 Gallery

Barplot: https://r-graph-gallery.com/barplot.html
Lollipop plot: https://r-graph-gallery.com/lollipop-plot.html

7.4 Multivariate Correlation Matrices

Used for: association between multiple numeric variables

For correlation matrices, I do the following:

I use the lab’s cor.table() function (with type = "manuscript") from the petersenlab package to create a correlation matrix.
I save the correlation matrix to a .csv file.
1. For example: https://research-git.uiowa.edu/PetersenLab/SRS/SRS-SelfRegulationDevelopment/-/blob/master/Analyses/selfRegulationDevelopment.Rmd#self-regulation-measures
I open the .csv file in Excel and create the table in Excel that can be copied and pasted to Word/Powerpoint/etc.

7.4.1 Correlograms

corrplot package: https://cran.r-project.org/web/packages/corrplot/vignettes/corrplot-intro.html
corrgram package: https://cran.r-project.org/web/packages/corrgram/vignettes/corrgram_examples.html

7.4.1.1 Gallery

https://r-graph-gallery.com/correlogram.html

7.4.2 Pairs panels

psych package: https://personality-project.org/r/psych/help/pairs.panels.html

I depict examples of correlograms and pairs panels here: https://isaactpetersen.github.io/Principles-Psychological-Assessment/factor-analysis-pca.html#sec-correlations-factorAnalysis

7.5 Path Diagrams

Used for: SEM/CFA/path analysis

If you are just trying to visualize the results of a SEM model fitted using the lavaan package, I recommend the semPlot, lavaanPlot, or lavaangui packages in R. You can access a web application version of lavaangui here: https://lavaangui.org. You can see examples of semPlot here: http://sachaepskamp.com/semPlot/examples (archived at: https://perma.cc/W2V4-C9C8). You can see examples of lavaanPlot here: https://lavaanplot.alexlishinski.com/articles/intro_to_lavaanplot (archived at: https://perma.cc/ARZ7-MV24). You can see examples of lavaangui here: https://doi.org/10.1080/10705511.2024.2420678. You can see examples of my implementation here: https://isaactpetersen.github.io/Principles-Psychological-Assessment/structural-equation-modeling.html#sec-semModelPathDiagram-sem

If you are trying to create a figure for a paper or poster, you might want something that you can draw and customize yourself. I use Adobe Illustrator for hand-drawn figures.

You can look at various options below:

semPlot package: https://doi.org/10.32614/CRAN.package.semPlot
lavaanPlot package: https://doi.org/10.32614/CRAN.package.lavaanPlot
lavaangui package: https://doi.org/10.32614/CRAN.package.lavaangui
Adobe Illustrator
Draw.io: https://www.drawio.com (formerly https://www.diagrams.net)
Onyx: https://onyx-sem.com
yworks: https://www.yworks.com
Microsoft Visio: https://www.microsoft.com/en-us/microsoft-365/visio/flowchart-software
Microsoft Powerpoint
AMOS
Warppls
Graphviz: https://graphviz.org
- has an R port—this is what we use for our study flowchart via the DiagrammeR package: https://rich-iannone.github.io/DiagrammeR/index.html
https://app.diagrams.net
https://github.com/jgraph/drawio-desktop/releases

7.6 Interactive

Gallery: https://r-graph-gallery.com/interactive-charts.html

7.7 Animation

Gallery: https://r-graph-gallery.com/animation.html

7.8 3D

Gallery: https://r-graph-gallery.com/3d.html

8 Color Palettes

8.1 Sequential

https://colorbrewer2.org/#type=sequential&scheme=BuGn&n=3
viridis, cividis, etc.: https://cran.r-project.org/web/packages/viridis/vignettes/intro-to-viridis.html

8.2 Diverging

https://colorbrewer2.org/#type=diverging&scheme=BrBG&n=3

8.3 Qualitative/Categorical

https://colorbrewer2.org/#type=qualitative&scheme=Accent&n=3
Polychrome package: https://stackoverflow.com/a/62939405 (archived at https://perma.cc/3HWM-MMFS)
pals package: https://stackoverflow.com/a/41230685 (archived at https://perma.cc/WH56-HMVD)

Color palettes for color-blindness:

Safe palette from the rcartocolor package: https://stackoverflow.com/a/56066712 (archived at https://perma.cc/WUH5-F4Z7)
Okabe Ito scale: https://stackoverflow.com/a/56066712 (archived at https://perma.cc/WUH5-F4Z7)

8.3.1 8 Colors

Code

# From here: https://github.com/clauswilke/colorblindr/blob/master/R/palettes.R
# Two color palettes taken from the article ["Color Universal Design" by Okabe and Ito](https://web.archive.org/web/20210108233739/http://jfly.iam.u-tokyo.ac.jp/color/)
# The variant `palette_OkabeIto` contains a gray color, while `palette_OkabeIto_black` contains black instead

palette_OkabeIto <- c("#E69F00", "#56B4E9", "#009E73", "#F0E442", "#0072B2", "#D55E00", "#CC79A7", "#999999")

pie(rep(1, 8), col = palette_OkabeIto)

Code

palette_OkabeIto_black <- c("#E69F00", "#56B4E9", "#009E73", "#F0E442", "#0072B2", "#D55E00", "#CC79A7", "#000000")

pie(rep(1, 8), col = palette_OkabeIto_black)

8.3.2 12 Colors

Safe palette from the rcartocolor package: https://stackoverflow.com/a/56066712 (archived at https://perma.cc/WUH5-F4Z7)

Code

#from: scales::show_col(carto_pal(12, "Safe"))

c12 <- c(
  "#88CCEE",
  "#CC6677",
  "#DDCC77",
  "#117733",
  "#332288",
  "#AA4499",
  "#44AA99",
  "#999933",
  "#882255",
  "#661100",
  "#6699CC",
  "#888888"
)

pie(rep(1, 12), col = c12)

8.3.3 25 Colors

https://stackoverflow.com/a/9568659 (archived at https://perma.cc/5ALZ-3AQD)

Code

c25 <- c(
  "dodgerblue2", "#E31A1C", # red
  "green4",
  "#6A3D9A", # purple
  "#FF7F00", # orange
  "black", "gold1",
  "skyblue2", "#FB9A99", # lt pink
  "palegreen2",
  "#CAB2D6", # lt purple
  "#FDBF6F", # lt orange
  "gray70", "khaki2",
  "maroon", "orchid1", "deeppink1", "blue1", "steelblue4",
  "darkturquoise", "green1", "yellow4", "yellow3",
  "darkorange4", "brown"
)

pie(rep(1, 25), col = c25)

8.3.4 36 Colors

Code

# from: Polychrome::palette36.colors(36)

c36 <- c("#5A5156","#E4E1E3","#F6222E","#FE00FA","#16FF32","#3283FE","#FEAF16","#B00068","#1CFFCE","#90AD1C","#2ED9FF","#DEA0FD","#AA0DFE","#F8A19F","#325A9B","#C4451C","#1C8356","#85660D","#B10DA1","#FBE426","#1CBE4F","#FA0087",
"#FC1CBF","#F7E1A0","#C075A6","#782AB6","#AAF400","#BDCDFF","#822E1C","#B5EFB5","#7ED7D1","#1C7F93","#D85FF7","#683B79","#66B0FF","#3B00FB")

pie(rep(1, 36), col = c36)

8.4 Maps

https://github.com/thomasp85/scico

9 Session Info

Code

sessionInfo()

R version 4.5.2 (2025-10-31)
Platform: x86_64-pc-linux-gnu
Running under: Ubuntu 24.04.3 LTS

Matrix products: default
BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3 
LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.26.so;  LAPACK version 3.12.0

locale:
 [1] LC_CTYPE=C.UTF-8       LC_NUMERIC=C           LC_TIME=C.UTF-8       
 [4] LC_COLLATE=C.UTF-8     LC_MONETARY=C.UTF-8    LC_MESSAGES=C.UTF-8   
 [7] LC_PAPER=C.UTF-8       LC_NAME=C              LC_ADDRESS=C          
[10] LC_TELEPHONE=C         LC_MEASUREMENT=C.UTF-8 LC_IDENTIFICATION=C   

time zone: UTC
tzcode source: system (glibc)

attached base packages:
[1] grid      stats     graphics  grDevices utils     datasets  methods  
[8] base     

other attached packages:
 [1] ggpubr_0.6.2       ggthemes_5.2.0     viridis_0.6.5      viridisLite_0.4.2 
 [5] ggExtra_0.11.0     reshape2_1.4.5     RColorBrewer_1.1-3 plyr_1.8.9        
 [9] reshape_0.8.10     ggplot2_4.0.1      ellipse_0.5.0      petersenlab_1.2.2 

loaded via a namespace (and not attached):
 [1] tidyselect_1.2.1  psych_2.5.6       dplyr_1.1.4       farver_2.1.2     
 [5] S7_0.2.1          fastmap_1.2.0     promises_1.5.0    digest_0.6.39    
 [9] rpart_4.1.24      mime_0.13         lifecycle_1.0.4   cluster_2.1.8.1  
[13] magrittr_2.0.4    compiler_4.5.2    rlang_1.1.6       Hmisc_5.2-4      
[17] tools_4.5.2       yaml_2.3.12       data.table_1.18.0 ggsignif_0.6.4   
[21] knitr_1.51        labeling_0.4.3    htmlwidgets_1.6.4 mnormt_2.1.1     
[25] abind_1.4-8       miniUI_0.1.2      foreign_0.8-90    withr_3.0.2      
[29] purrr_1.2.0       nnet_7.3-20       stats4_4.5.2      lavaan_0.6-21    
[33] xtable_1.8-4      colorspace_2.1-2  scales_1.4.0      MASS_7.3-65      
[37] cli_3.6.5         mvtnorm_1.3-3     rmarkdown_2.30    reformulas_0.4.3 
[41] generics_0.1.4    otel_0.2.0        rstudioapi_0.17.1 minqa_1.2.8      
[45] DBI_1.2.3         stringr_1.6.0     splines_4.5.2     parallel_4.5.2   
[49] base64enc_0.1-3   mitools_2.4       vctrs_0.6.5       boot_1.3-32      
[53] Matrix_1.7-4      carData_3.0-5     jsonlite_2.0.0    car_3.1-3        
[57] rstatix_0.7.3     Formula_1.2-5     htmlTable_2.4.3   tidyr_1.3.2      
[61] glue_1.8.0        nloptr_2.2.1      stringi_1.8.7     gtable_0.3.6     
[65] later_1.4.4       quadprog_1.5-8    lme4_1.1-38       tibble_3.3.0     
[69] pillar_1.11.1     htmltools_0.5.9   R6_2.6.1          Rdpack_2.6.4     
[73] mix_1.0-13        evaluate_1.0.5    shiny_1.12.1      pbivnorm_0.6.0   
[77] lattice_0.22-7    rbibutils_2.4     backports_1.5.0   broom_1.0.11     
[81] httpuv_1.6.16     Rcpp_1.1.0        gridExtra_2.3     nlme_3.1-168     
[85] checkmate_2.3.3   mgcv_1.9-3        xfun_0.55         pkgconfig_2.0.3

Footnotes

Ask me to give you access.↩︎

Reuse

CC BY 4.0

--- title: "Figures in R" --- # Picking a Chart Type <https://www.data-to-viz.com> # Gallery <https://www.r-graph-gallery.com> # Resources for Learning `R` Syntax for Figures <https://www.statmethods.net/graphs/index.html> <http://www.cookbook-r.com/Graphs/> # Petersen Lab Examples <https://research-git.uiowa.edu/PetersenLab/R-Plotting/-/tree/main/Analyses> # Font You can download the lab fonts for figures here[^1]: <https://drive.google.com/drive/u/0/folders/1fqlrnEe7NFnWZoIrsHmr8ulDS4nhs-H3> ## Preamble ### Install Libraries ```{r} #install.packages("remotes") #remotes::install_github("DevPsyLab/petersenlab") ``` ### Load Libraries ```{r} library("petersenlab") library("ellipse") library("ggplot2") library("grid") library("reshape") library("plyr") library("RColorBrewer") library("reshape2") library("ggExtra") library("viridis") library("ggthemes") library("ggpubr") ``` ## Simulate Data ```{r} set.seed(52242) n <- 1000 predictor <- rbeta(n, 1.5, 5) * 100 outcome <- predictor + rnorm(n, mean = 0, sd = 20) + 50 number <- sample(1:1000, replace = TRUE) predictorOverplot <- sample(1:50, n, replace = TRUE) outcomeOverplot <- predictorOverplot + sample(1:75, n, replace = TRUE) df <- data.frame(predictor = predictor, outcome = outcome, predictorOverplot = predictorOverplot, outcomeOverplot = outcomeOverplot) df[sample(1:n, size = 10), "predictor"] <- NA df[sample(1:n, size = 10), "outcome"] <- NA df[sample(1:n, size = 10), "predictorOverplot"] <- NA df[sample(1:n, size = 10), "outcomeOverplot"] <- NA ``` ## Line ```{r} plot.new() lines( x = seq(from = -10, to = 10, length.out = 100), y = seq(from = -25, to = 25, length.out = 100)) ``` ## Curve ```{r} curve(x^3 - 3*x, from = -2, to = 2) curve(x^2 - 2, add = TRUE, col = "violet") ``` ## Basic Scatterplot ### Base R ```{r} plot(outcome ~ predictor, data = df) plot(df$predictor, df$outcome) ``` #### Best-fit line ```{r} plot(outcome ~ predictor, data = df) abline(lm(outcome ~ predictor, data = df), col = "red") #regression line (y~x) ``` #### Best-fit line with correlation coefficient ```{r} plot(outcome ~ predictor, data = df) abline(lm(outcome ~ predictor, data = df), col = "red") #regression line (y~x) addText(x = df$predictor, y = df$outcome) ``` #### Loess line ```{r} plot(outcome ~ predictor, data = df) lines(loess.smooth(df$predictor, df$outcome)) #loess line (x,y) ``` ### `ggplot2` ```{r} ggplot(df, aes(x = predictor, y = outcome)) + geom_point() + theme_classic() ``` #### Best-fit line ```{r} ggplot(df, aes(x = predictor, y = outcome)) + geom_point() + stat_smooth(method = "lm", formula = y ~ x) + theme_classic() ``` #### Best-fit line with correlation coefficient ```{r} ggplot(df, aes(x = predictor, y = outcome)) + geom_point() + stat_smooth(method = "lm", formula = y ~ x) + stat_cor( cor.coef.name = "r", p.accuracy = 0.001, r.accuracy = 0.01) + theme_classic() ``` #### Loess line ```{r} ggplot(df, aes(x = predictor, y = outcome)) + geom_point() + stat_smooth(method = "loess", formula = y ~ x) + theme_classic() ``` ## Change Plot Style ### Change Theme ```{r} basePlot <- ggplot(df, aes(x = predictor, y = outcome)) + geom_point() ``` #### Default Theme ```{r} basePlot ``` #### Grayscale: `theme_gray()` ```{r} basePlot + theme_gray() + theme(text = element_text(family = "Gotham")) ``` #### Black-and-White: `theme_bw()` ```{r} basePlot + theme_bw() + theme(text = element_text(family = "Gotham")) ``` #### Line Drawing: `theme_linedraw()` A theme with only black lines of various widths on white backgrounds, reminiscent of a line drawing. Note that this theme has some very thin lines (<< 1 pt) which some journals may refuse. ```{r} basePlot + theme_linedraw() + theme(text = element_text(family = "Gotham")) ``` #### Light: `theme_light()` ```{r} basePlot + theme_light() + theme(text = element_text(family = "Gotham")) ``` #### Dark: `theme_dark()` ```{r} basePlot + theme_dark() + theme(text = element_text(family = "Gotham")) ``` #### Minimal: `theme_minimal()` ```{r} basePlot + theme_minimal() + theme(text = element_text(family = "Gotham")) ``` #### Classic: `theme_classic()` ```{r} basePlot + theme_classic() + theme(text = element_text(family = "Gotham")) ``` #### A Completely Empty Theme: `theme_void()` ```{r} basePlot + theme_void() + theme(text = element_text(family = "Gotham")) ``` #### Visual Unit Tests: `theme_test()` ```{r} basePlot + theme_test() + theme(text = element_text(family = "Gotham")) ``` #### Edward Tufte: `theme_tufte()` Theme based on Edward Tufte. ```{r} basePlot + theme_tufte() ``` #### Wall Street Journal: `theme_wsj()` Theme based on the publication, the Wall Street Journal. ```{r} basePlot + theme_wsj() ``` #### FiveThirtyEight: `theme_fivethirtyeight()` Theme based on the publication, FiveThirtyEight. ```{r} basePlot + theme_fivethirtyeight() ``` #### The Economist: `theme_economist()` Theme based on the publication, The Economist. ```{r} basePlot + theme_economist() ``` #### Stephen Few: `theme_few()` Theme based on the rules and examples from Stephen Few's *Show Me the Numbers* and "Practical Rules for Using Color in Charts". ```{r} basePlot + theme_few() ``` ## Add Marginal Distributions {#sec-marginalDistributions} ```{r} scatterplot <- ggplot(df, aes(x = predictor, y = outcome)) + geom_point() + theme_classic() + theme(text = element_text(family = "Gotham")) ``` ### Density Plot ```{r} densityMarginal <- ggMarginal(scatterplot, type = "density", xparams = list(fill = "gray"), yparams = list(fill = "gray")) ``` ```{r} print(densityMarginal, newpage = TRUE) ``` ### Histogram ```{r} histogramMarginal <- ggMarginal(scatterplot, type = "histogram", xparams = list(fill = "gray"), yparams = list(fill = "gray")) ``` ```{r} print(histogramMarginal, newpage = TRUE) ``` ### Boxplot ```{r} boxplotMarginal <- ggMarginal(scatterplot, type = "boxplot", xparams = list(fill = "gray"), yparams = list(fill = "gray")) ``` ```{r} print(boxplotMarginal, newpage = TRUE) ``` ### Violin Plot ```{r} violinMarginal <- ggMarginal(scatterplot, type = "violin", xparams = list(fill = "gray"), yparams = list(fill = "gray")) ``` ```{r} print(violinMarginal, newpage = TRUE) ``` ### Density Plot and Histogram ```{r} densigramMarginal <- ggMarginal(scatterplot, type = "densigram", xparams = list(fill = "gray"), yparams = list(fill = "gray")) ``` ```{r} print(densigramMarginal, newpage = TRUE) ``` ## Ellipse ### Basic Ellipse ```{r} ggplot(df, aes(x = predictor, y = outcome)) + geom_point() + stat_ellipse(alpha = 0.4, level = 0.95, geom = "polygon", fill = "red", color = "red") + theme_classic() + theme(text = element_text(family = "Gotham")) ``` ### Align Coordinates ```{r} ggplot(df, aes(x = predictor, y = outcome)) + geom_point() + stat_ellipse(alpha = 0.4, level = 0.95, geom = "polygon", fill = "red", color = "red") + scale_x_continuous(expand = c(0, 0)) + scale_y_continuous(expand = c(0, 0)) + coord_fixed(ratio = (max(predictor, na.rm = TRUE) - min(predictor, na.rm = TRUE))/(max(outcome, na.rm = TRUE) - min(outcome, na.rm = TRUE)), xlim = c(0, max(predictor, na.rm = TRUE)), ylim = c(0, max(outcome, na.rm = TRUE))) + theme_classic() + theme(text = element_text(family = "Gotham")) ``` ### Reduce Dot Size ```{r} ggplot(df, aes(x = predictor, y = outcome)) + geom_point(size = 0.5) + stat_ellipse(alpha = 0.4, level = 0.95, geom = "polygon", fill = "red", color = "red") + scale_x_continuous(expand = c(0, 0)) + scale_y_continuous(expand = c(0, 0)) + coord_fixed(ratio = (max(predictor, na.rm = TRUE) - min(predictor, na.rm = TRUE))/(max(outcome, na.rm = TRUE) - min(outcome, na.rm = TRUE)), xlim = c(0, max(predictor, na.rm = TRUE)), ylim = c(0, max(outcome, na.rm = TRUE))) + theme_classic() + theme(text = element_text(family = "Gotham")) ``` ### Transparency ```{r} ggplot(df, aes(x = predictor, y = outcome)) + geom_point(alpha = 0.3) + stat_ellipse(alpha = 0.4, level = 0.95, geom = "polygon", fill = "red", color = "red") + scale_x_continuous(expand = c(0, 0)) + scale_y_continuous(expand = c(0, 0)) + coord_fixed(ratio = (max(predictor, na.rm = TRUE) - min(predictor, na.rm = TRUE))/(max(outcome, na.rm = TRUE) - min(outcome, na.rm = TRUE)), xlim = c(0, max(predictor, na.rm = TRUE)), ylim = c(0, max(outcome, na.rm = TRUE))) + theme_classic() + theme(text = element_text(family = "Gotham")) ``` ## Bubble Chart ### Basic Bubble Chart ```{r} ggplot(df, aes(x = predictorOverplot, y = outcomeOverplot)) + geom_count(aes(size = ..n..)) + scale_size_area() + theme_classic() + theme(text = element_text(family = "Gotham")) ``` ### Specify Sizes ```{r} ggplot(df, aes(x = predictorOverplot, y = outcomeOverplot)) + geom_count(aes(size = ..n..)) + scale_size_continuous(breaks = c(1, 2, 3, 4), range = c(1, 7)) + theme_classic() + theme(text = element_text(family = "Gotham")) ``` ## 2-Dimensional Density ```{r} ggplot(df, aes(x = predictor, y = outcome)) + stat_density_2d(aes(fill = ..density..), geom = "raster", contour = FALSE) + scale_x_continuous(expand = c(0, 0)) + scale_y_continuous(expand = c(0, 0)) + scale_fill_viridis() + theme( legend.position = "none", text = element_text(family = "Gotham") ) ``` ## Combined Ellipse and Bubble Chart ### `ggplot2` ```{r} ggplot(df, aes(x = predictorOverplot, y = outcomeOverplot)) + geom_count(alpha = .6, color = rgb(0,0,.7,.5)) + scale_size_continuous(breaks = c(1, 2, 3, 4), range = c(1, 7)) + stat_smooth(method = "loess", se = TRUE, color = "green") + stat_smooth(method = "lm") + stat_ellipse(alpha = 0.4, level = 0.95, geom = "polygon", fill = "red", color = "red") + scale_x_continuous(expand = c(0, 0)) + scale_y_continuous(expand = c(0, 0)) + coord_fixed(ratio = (max(predictorOverplot, na.rm = TRUE) - min(predictorOverplot, na.rm = TRUE))/(max(outcomeOverplot, na.rm = TRUE) - min(outcomeOverplot, na.rm = TRUE)), xlim = c(0, max(predictorOverplot, na.rm = TRUE)), ylim = c(0, max(outcomeOverplot, na.rm = TRUE))) + theme_classic() + theme(text = element_text(family = "Gotham")) ggplot(df, aes(x = predictorOverplot, y = outcomeOverplot)) + geom_count(alpha = .6, color = rgb(0,0,.7,.5)) + scale_size_continuous(breaks = c(1, 2, 3, 4), range = c(1, 7)) + stat_smooth(method = "loess", se = TRUE, color = "green") + stat_smooth(method = "lm") + stat_ellipse(color = "red", size = 1.5) + scale_x_continuous(expand = c(0, 0)) + scale_y_continuous(expand = c(0, 0)) + coord_fixed(ratio = (max(predictorOverplot, na.rm = TRUE) - min(predictorOverplot, na.rm = TRUE))/(max(outcomeOverplot, na.rm = TRUE) - min(outcomeOverplot, na.rm = TRUE)), xlim = c(0, max(predictorOverplot, na.rm = TRUE)), ylim = c(0, max(outcomeOverplot, na.rm = TRUE))) + theme_classic() + theme(text = element_text(family = "Gotham")) ``` ### Other implementation From: <https://stats.stackexchange.com/questions/7899/complex-regression-plot-in-r> #### `ggplot2` ```{r} df$x <- df$predictorOverplot df$y <- df$outcomeOverplot xc <- with(df, xyTable(x, y)) df2 <- cbind.data.frame(x = xc$x, y = xc$y, number = xc$number) df2$n <- cut(df2$number, c(0,1.5,2.5,Inf), labels = c(1,2,4)) df.ell <- as.data.frame(with(df, ellipse(cor(df$x, df$y, use = "pairwise.complete.obs"), scale = c(sd(df$x, na.rm = TRUE), sd(df$y, na.rm = TRUE)), centre = c(mean(df$x, na.rm = TRUE), mean(df$y, na.rm = TRUE)), level = .95))) ggplot(data = na.omit(df2), aes(x = x, y = y)) + geom_point(aes(size = n), alpha = .6, color = rgb(0,0,.7,.5)) + stat_smooth(data = df, method = "loess", se = FALSE, color = "green") + stat_smooth(data = df, method = "lm", col = "red") + geom_path(data = df.ell, colour = "green", size = 1) + coord_cartesian(xlim = c(-1,60), ylim = c(-1,130)) ``` #### Base `R` ```{r} do.it <- function(df, type="confidence", ...) { require(ellipse) lm0 <- lm(y ~ x, data=df) xc <- with(df, xyTable(x, y)) df.new <- data.frame(x = seq(min(df$x), max(df$x), 0.1)) pred.ulb <- predict(lm0, df.new, interval = type) pred.lo <- predict(loess(y ~ x, data = df), df.new) plot(xc$x, xc$y, cex = xc$number*1/4, xlab = "x", ylab = "y", ...) #change number*X to change dot size abline(lm0, col = "red") lines(df.new$x, pred.lo, col="green", lwd = 2) lines(df.new$x, pred.ulb[,"lwr"], lty = 2, col = "red") lines(df.new$x, pred.ulb[,"upr"], lty = 2, col = "red") lines(ellipse(cor(df$x, df$y), scale=c(sd(df$x),sd(df$y)), centre = c(mean(df$x), mean(df$y)), level = .95), lwd = 2, col = "green") invisible(lm0) } df3 <- na.omit(df[sample(nrow(df), nrow(df), rep = TRUE),]) df3$x <- df3$predictorOverplot df3$y <- df3$outcomeOverplot do.it(df3, pch = 19, col = rgb(0,0,.7,.5)) ``` ## Visually-Weighted Regression ### Default ```{r} #| results: false vwReg(outcome ~ predictor, data = df) ``` ### Shade ```{r} #| results: false vwReg(outcome ~ predictor, data = df, shade = TRUE, spag = FALSE, show.lm = TRUE, show.CI = TRUE, bw = FALSE, B = 1000, quantize = "continuous") vwReg(outcome ~ predictor, data = df, shade = TRUE, spag = FALSE, show.lm = TRUE, show.CI = TRUE, bw = FALSE, B = 1000, quantize = "SD") ``` ### Spaghetti ```{r} vwReg(outcome ~ predictor, data = df, shade = FALSE, spag = TRUE, show.lm = TRUE, show.CI = TRUE, bw = FALSE, B = 1000) vwReg(outcome ~ predictor, data = df, shade = FALSE, spag = TRUE, show.lm = FALSE, show.CI = FALSE, bw = FALSE, B = 1000) ``` ### Black/white ```{r} #| results: false vwReg(outcome ~ predictor, data = df, shade = TRUE, spag = FALSE, show.lm = TRUE, show.CI = TRUE, bw = TRUE, B = 1000, quantize = "continuous") vwReg(outcome ~ predictor, data = df, shade = TRUE, spag = FALSE, show.lm = TRUE, show.CI = TRUE, bw = TRUE, B = 1000, quantize = "SD") vwReg(outcome ~ predictor, data = df, shade = FALSE, spag = TRUE, show.lm = TRUE, show.CI = TRUE, bw = TRUE, B = 1000, quantize = "SD") ``` # Graphic Design Principles for Data Visualization <https://www.data-to-viz.com/caveats.html> # Types of Plots ## Univariate Distribution Used for: distribution of one numeric variable ### Gallery - Violin chart: <https://r-graph-gallery.com/violin.html> - Density chart: <https://r-graph-gallery.com/density-plot.html> - Histogram: <https://r-graph-gallery.com/histogram.html> - Boxplot: <https://r-graph-gallery.com/boxplot.html> - Ridgeline chart: <https://r-graph-gallery.com/ridgeline-plot.html> ## Bivariate Scatterplots Used for: association between two numeric variables ### Base `R` ``` plot(x, y) ``` ### ggplot2 package <http://www.cookbook-r.com/Graphs/Scatterplots_(ggplot2)/> ``` ggplot(data, aes(x, y)) + geom_point() ``` ### Gallery - Scatterplot: <https://r-graph-gallery.com/scatterplot.html> - Bubble plot: <https://r-graph-gallery.com/bubble-chart.html> - 2D density chart: <https://r-graph-gallery.com/2d-density-chart.html> - Heatmap: <https://r-graph-gallery.com/heatmap.html> ### Add lines - Line chart: <https://r-graph-gallery.com/line-plot.html> - Connected scatterplot: <https://r-graph-gallery.com/connected-scatterplot.html> - Visually-weighted regression: <https://www.nicebread.de/visually-weighted-watercolor-plots-new-variants-please-vote/> - Use the `vwReg()` function from the `petersenlab` package: <https://github.com/DevPsyLab/petersenlab/blob/main/R/vwReg.R> ### Area - Area chart: <https://r-graph-gallery.com/area-chart.html> - Stacked area chart: <https://r-graph-gallery.com/stacked-area-graph.html> - Streamgraph: <https://r-graph-gallery.com/streamgraph.html> ## Bivariate Barplots Used for: association between one categorical variable and one numeric variable (or for depicting the frequency of categories of a categorical variable) ### Gallery - Barplot: <https://r-graph-gallery.com/barplot.html> - Lollipop plot: <https://r-graph-gallery.com/lollipop-plot.html> ## Multivariate Correlation Matrices Used for: association between multiple numeric variables For correlation matrices, I do the following: 1. I use the lab's `cor.table()` function (with `type = "manuscript"`) from the `petersenlab` package to create a correlation matrix. 1. I save the correlation matrix to a `.csv` file. a. For example: <https://research-git.uiowa.edu/PetersenLab/SRS/SRS-SelfRegulationDevelopment/-/blob/master/Analyses/selfRegulationDevelopment.Rmd#self-regulation-measures> 1. I open the .csv file in Excel and create the table in Excel that can be copied and pasted to Word/Powerpoint/etc. ### Correlograms - `corrplot` package: <https://cran.r-project.org/web/packages/corrplot/vignettes/corrplot-intro.html> - `corrgram` package: <https://cran.r-project.org/web/packages/corrgram/vignettes/corrgram_examples.html> #### Gallery <https://r-graph-gallery.com/correlogram.html> ### Pairs panels `psych` package: <https://personality-project.org/r/psych/help/pairs.panels.html> I depict examples of correlograms and pairs panels here: <https://isaactpetersen.github.io/Principles-Psychological-Assessment/factor-analysis-pca.html#sec-correlations-factorAnalysis> ## Path Diagrams {#sec-pathDiagrams} Used for: SEM/CFA/path analysis If you are just trying to visualize the results of a SEM model fitted using the `lavaan` package, I recommend the [`semPlot`](https://doi.org/10.32614/CRAN.package.semPlot), [`lavaanPlot`](https://doi.org/10.32614/CRAN.package.lavaanPlot), or [`lavaangui`](https://doi.org/10.32614/CRAN.package.lavaangui) packages in `R`. You can access a web application version of [`lavaangui`](https://doi.org/10.32614/CRAN.package.lavaangui) here: <https://lavaangui.org>. You can see examples of `semPlot` here: <http://sachaepskamp.com/semPlot/examples> (archived at: <https://perma.cc/W2V4-C9C8>). You can see examples of `lavaanPlot` here: <https://lavaanplot.alexlishinski.com/articles/intro_to_lavaanplot> (archived at: <https://perma.cc/ARZ7-MV24>). You can see examples of `lavaangui` here: <https://doi.org/10.1080/10705511.2024.2420678>. You can see examples of my implementation here: <https://isaactpetersen.github.io/Principles-Psychological-Assessment/structural-equation-modeling.html#sec-semModelPathDiagram-sem> If you are trying to create a figure for a paper or poster, you might want something that you can draw and customize yourself. I use Adobe Illustrator for hand-drawn figures. You can look at various options below: - `semPlot` package: <https://doi.org/10.32614/CRAN.package.semPlot> - `lavaanPlot` package: <https://doi.org/10.32614/CRAN.package.lavaanPlot> - `lavaangui` package: <https://doi.org/10.32614/CRAN.package.lavaangui> - `Adobe Illustrator` - `Draw.io`: <https://www.drawio.com> (formerly <https://www.diagrams.net>) - `Onyx`: <https://onyx-sem.com> - `yworks`: <https://www.yworks.com> - `Microsoft Visio`: <https://www.microsoft.com/en-us/microsoft-365/visio/flowchart-software> - `Microsoft Powerpoint` - `AMOS` - `Warppls` - `Graphviz`: <https://graphviz.org> - has an `R` port—this is what we use for our study flowchart via the `DiagrammeR` package: <https://rich-iannone.github.io/DiagrammeR/index.html> - <https://app.diagrams.net> - <https://github.com/jgraph/drawio-desktop/releases> ## Interactive Gallery: <https://r-graph-gallery.com/interactive-charts.html> ## Animation Gallery: <https://r-graph-gallery.com/animation.html> ## 3D Gallery: <https://r-graph-gallery.com/3d.html> # Color Palettes {#sec-colorPalettes} - <https://colorbrewer2.org> - <https://sites.stat.columbia.edu/tzheng/files/Rcolor.pdf> (archived at <https://perma.cc/8SYT-UC5U>) ## Sequential - <https://colorbrewer2.org/#type=sequential&scheme=BuGn&n=3> - viridis, cividis, etc.: <https://cran.r-project.org/web/packages/viridis/vignettes/intro-to-viridis.html> ## Diverging - <https://colorbrewer2.org/#type=diverging&scheme=BrBG&n=3> ## Qualitative/Categorical - <https://colorbrewer2.org/#type=qualitative&scheme=Accent&n=3> - `Polychrome` package: <https://stackoverflow.com/a/62939405> (archived at <https://perma.cc/3HWM-MMFS>) - `pals` package: <https://stackoverflow.com/a/41230685> (archived at <https://perma.cc/WH56-HMVD>) Color palettes for color-blindness: - `Safe` palette from the `rcartocolor` package: <https://stackoverflow.com/a/56066712> (archived at <https://perma.cc/WUH5-F4Z7>) - Okabe Ito scale: <https://stackoverflow.com/a/56066712> (archived at <https://perma.cc/WUH5-F4Z7>) ### 8 Colors ```{r} # From here: https://github.com/clauswilke/colorblindr/blob/master/R/palettes.R # Two color palettes taken from the article ["Color Universal Design" by Okabe and Ito](https://web.archive.org/web/20210108233739/http://jfly.iam.u-tokyo.ac.jp/color/) # The variant `palette_OkabeIto` contains a gray color, while `palette_OkabeIto_black` contains black instead palette_OkabeIto <- c("#E69F00", "#56B4E9", "#009E73", "#F0E442", "#0072B2", "#D55E00", "#CC79A7", "#999999") pie(rep(1, 8), col = palette_OkabeIto) palette_OkabeIto_black <- c("#E69F00", "#56B4E9", "#009E73", "#F0E442", "#0072B2", "#D55E00", "#CC79A7", "#000000") pie(rep(1, 8), col = palette_OkabeIto_black) ``` ### 12 Colors `Safe` palette from the `rcartocolor` package: <https://stackoverflow.com/a/56066712> (archived at <https://perma.cc/WUH5-F4Z7>) ```{r} #from: scales::show_col(carto_pal(12, "Safe")) c12 <- c( "#88CCEE", "#CC6677", "#DDCC77", "#117733", "#332288", "#AA4499", "#44AA99", "#999933", "#882255", "#661100", "#6699CC", "#888888" ) pie(rep(1, 12), col = c12) ``` ### 25 Colors <https://stackoverflow.com/a/9568659> (archived at <https://perma.cc/5ALZ-3AQD>) ```{r} c25 <- c( "dodgerblue2", "#E31A1C", # red "green4", "#6A3D9A", # purple "#FF7F00", # orange "black", "gold1", "skyblue2", "#FB9A99", # lt pink "palegreen2", "#CAB2D6", # lt purple "#FDBF6F", # lt orange "gray70", "khaki2", "maroon", "orchid1", "deeppink1", "blue1", "steelblue4", "darkturquoise", "green1", "yellow4", "yellow3", "darkorange4", "brown" ) pie(rep(1, 25), col = c25) ``` ### 36 Colors ```{r} # from: Polychrome::palette36.colors(36) c36 <- c("#5A5156","#E4E1E3","#F6222E","#FE00FA","#16FF32","#3283FE","#FEAF16","#B00068","#1CFFCE","#90AD1C","#2ED9FF","#DEA0FD","#AA0DFE","#F8A19F","#325A9B","#C4451C","#1C8356","#85660D","#B10DA1","#FBE426","#1CBE4F","#FA0087", "#FC1CBF","#F7E1A0","#C075A6","#782AB6","#AAF400","#BDCDFF","#822E1C","#B5EFB5","#7ED7D1","#1C7F93","#D85FF7","#683B79","#66B0FF","#3B00FB") pie(rep(1, 36), col = c36) ``` ## Maps - <https://github.com/thomasp85/scico> # Session Info ```{r} #| code-fold: true sessionInfo() ``` [^1]: Ask me to give you access.

1 Picking a Chart Type

2 Gallery

3 Resources for Learning R Syntax for Figures

4 Petersen Lab Examples

5 Font

5.1 Preamble

5.1.1 Install Libraries

5.1.2 Load Libraries

5.2 Simulate Data

5.3 Line

5.4 Curve

5.5 Basic Scatterplot

5.5.1 Base R

5.5.1.1 Best-fit line

5.5.1.2 Best-fit line with correlation coefficient

5.5.1.3 Loess line

5.5.2 ggplot2

5.5.2.1 Best-fit line

5.5.2.2 Best-fit line with correlation coefficient

5.5.2.3 Loess line

5.6 Change Plot Style

5.6.1 Change Theme

5.6.1.1 Default Theme

5.6.1.2 Grayscale: theme_gray()

5.6.1.3 Black-and-White: theme_bw()

5.6.1.4 Line Drawing: theme_linedraw()

5.6.1.5 Light: theme_light()

5.6.1.6 Dark: theme_dark()

5.6.1.7 Minimal: theme_minimal()

5.6.1.8 Classic: theme_classic()

5.6.1.9 A Completely Empty Theme: theme_void()

5.6.1.10 Visual Unit Tests: theme_test()

5.6.1.11 Edward Tufte: theme_tufte()

5.6.1.12 Wall Street Journal: theme_wsj()

5.6.1.13 FiveThirtyEight: theme_fivethirtyeight()

5.6.1.14 The Economist: theme_economist()

5.6.1.15 Stephen Few: theme_few()

5.7 Add Marginal Distributions

5.7.1 Density Plot

5.7.2 Histogram

5.7.3 Boxplot

5.7.4 Violin Plot

5.7.5 Density Plot and Histogram

5.8 Ellipse

5.8.1 Basic Ellipse

5.8.2 Align Coordinates

5.8.3 Reduce Dot Size

5.8.4 Transparency

5.9 Bubble Chart

5.9.1 Basic Bubble Chart

5.9.2 Specify Sizes

5.10 2-Dimensional Density

5.11 Combined Ellipse and Bubble Chart

5.11.1 ggplot2

5.11.2 Other implementation

5.11.2.1 ggplot2

5.11.2.2 Base R

5.12 Visually-Weighted Regression

5.12.1 Default

5.12.2 Shade

5.12.3 Spaghetti

5.12.4 Black/white

6 Graphic Design Principles for Data Visualization

7 Types of Plots

7.1 Univariate Distribution

7.1.1 Gallery

7.2 Bivariate Scatterplots

7.2.1 Base R

7.2.2 ggplot2 package

7.2.3 Gallery

7.2.4 Add lines

7.2.5 Area

7.3 Bivariate Barplots

7.3.1 Gallery

7.4 Multivariate Correlation Matrices

7.4.1 Correlograms

7.4.1.1 Gallery

7.4.2 Pairs panels

7.5 Path Diagrams

7.6 Interactive

3 Resources for Learning `R` Syntax for Figures

5.5.2 `ggplot2`

5.6.1.2 Grayscale: `theme_gray()`

5.6.1.3 Black-and-White: `theme_bw()`

5.6.1.4 Line Drawing: `theme_linedraw()`

5.6.1.5 Light: `theme_light()`

5.6.1.6 Dark: `theme_dark()`

5.6.1.7 Minimal: `theme_minimal()`

5.6.1.8 Classic: `theme_classic()`

5.6.1.9 A Completely Empty Theme: `theme_void()`

5.6.1.10 Visual Unit Tests: `theme_test()`

5.6.1.11 Edward Tufte: `theme_tufte()`

5.6.1.12 Wall Street Journal: `theme_wsj()`

5.6.1.13 FiveThirtyEight: `theme_fivethirtyeight()`

5.6.1.14 The Economist: `theme_economist()`

5.6.1.15 Stephen Few: `theme_few()`

5.11.1 `ggplot2`

5.11.2.1 `ggplot2`

5.11.2.2 Base `R`

7.2.1 Base `R`