Why is the resolution on my geom_points so poor - r

Question

I came across an alternative to grouped bar charts in ggplot that Rebecca Barter posted on her blog and wanted to give it a try. It produces a slick Cleveland dot plot:
The code for my attempt follows:
ggplot() +
# remove axes and superfluous grids
theme_classic() +
theme(axis.ticks.y = element_blank(),
text = element_text(family = "Roboto Condensed"),
axis.text = element_text(size = rel(1.5)),
plot.title = element_text(size = 30, color = "#000000"),
plot.subtitle = element_text(size = 15, color = "#Ec111A"),
plot.caption = element_text(size = 15, color = "grey25"),
plot.margin = margin(20,20,20,20),
panel.background = element_rect(fill = "white"),
axis.line = element_blank(),
axis.text.x = element_text(vjust= + 15)) +
# add a dummy point for scaling purposes
geom_point(aes(x = 15, y = P),
size = 0, col = "white") +
# add the horizontal discipline lines
geom_hline(yintercept = 1:9, color = "grey80") +
# add a point for each male success rate
geom_point(aes(x = Male, y = P),
size = 15, col = "#00b0f0") +
# add a point for each female success rate
geom_point(aes(x = Female, y = P),
size = 15, col = "#Ec111A") +
geom_text(aes(x = Male, y = P,
label = paste0(round(Male, 1))),
col = "black", face="bold") +
# add the text (%) for each female success rate
geom_text(aes(x = Female, y = P,
label = paste0(round(Female, 1))),
col = "white", face="bold") +
# add a label above the first two points
geom_text(aes(x = x, y = y, label = label, col = label),
data.frame(x = c(21.8 - 0, 24.6 - 0), y = 7.5,
label = c("Male", "Female")), size = 6) +
scale_color_manual(values = c("#Ec111A", "#00b0f0"), guide = "none") +
# manually specify the x-axis
scale_x_continuous(breaks = c(0, 10, 20, 30),
labels = c("0%","10%", "20%", "30%")) +
# manually set the spacing above and below the plot
scale_y_discrete(expand = c(0.15, 0)) +
labs(
x = NULL,
y = NULL,
title= "Move Percentage By Gender",
subtitle = "What Percentage Of Moves Are Tops",
caption = "Takeaway: Males have fewer Tops and more Xs compared to Females.")
But my plot has very jagged (poor resolution points) and I can't figure out what's the cause.
Has anyone come across this problem and know how to fix it?

Answer 1

Saving and resolution depends on how you save and your graphics device. In other words... how are you saving your plot? Since it depends so much on your personal setup and parameters, your mileage will vary. One of the more dependable ways of saving plots from ggplot2 in R is to use ggsave(), where you can specify these parameters and maintain some consistency. Here is an example plot code:
ggplot(mtcars, aes(disp, mpg)) +
geom_point(size=10, color='red1') +
geom_text(aes(label=cyl), color='white')
This creates a plot similar to what you show using mtcars. If I copy and paste the graphic output directly from R or use export (I'm using RStudio) this is what you get:
Not sure if you can tell, but the edges are jagged and it does not look clean on close inspection. Definitely not OK for me. However, here's the same plot saved using ggsave():
ggsave('myplot.png', width = 9, height = 6)
You should be able to tell that it's a lot cleaner, because it is saved with a higher resolution. File size on the first is 9 KB, whereas it's 62 KB on the second.
In the end - just play with the settings on ggsave() and you should find some resolution that works for you. If you just input ggsave('myplotfile.png'), you'll get the width/height settings that match your viewport window in RStudio. You can get an idea of the aspect and size and adjust accordingly. One more point - be cautious that text does not scale the same as geoms, so your circles will increase in size differently than the text.