How to change the geom_point appearance? - r

Question

Using this data:
Data2 <- structure(list(year = c(2008L, 2009L, 2010L, 2011L, 2012L, 2013L,
2014L, 2015L, 2016L, 2017L, 2018L, 2019L, 2008L, 2009L, 2010L,
2011L, 2012L, 2013L, 2014L, 2015L, 2016L, 2017L, 2018L, 2019L,
2008L, 2009L, 2010L, 2011L, 2012L, 2013L, 2014L, 2015L, 2016L,
2017L, 2018L, 2019L), variable = c("var1", "var1", "var1", "var1",
"var1", "var1", "var1", "var1", "var1", "var1", "var1", "var1",
"var2", "var2", "var2", "var2", "var2", "var2", "var2", "var2",
"var2", "var2", "var2", "var2", "var3", "var3", "var3", "var3",
"var3", "var3", "var3", "var3", "var3", "var3", "var3", "var3"
), frequency = c(1L, 0L, 0L, 0L, 1L, 1L, 3L, 4L, 3L, 10L, 9L,
0L, 0L, 1L, 0L, 0L, 3L, 2L, 2L, 3L, 8L, 9L, 12L, 3L, 0L, 0L,
0L, 0L, 0L, 2L, 1L, 0L, 1L, 2L, 4L, 0L)), class = "data.frame", row.names = c(NA,
-36L))
I try to produce a plot like this one:
library(ggplot2)
ggplot(Data2, aes(year, variable, size = frequency, color = variable)) +
geom_point()+ theme(text = element_text(size = 18))
However I have some problem with the view of this graph:
How can I change the x axis to have all years, make bigger the geom_point (I tried the size but it is a fix option and all circles made the same) and from frequency show from 1 and not 0 cicles if exist?

Answer 1

An easy way to get years to show up is to turn them into factors. As far as starting your size scale at one: I'm not sure if you want to remove points representing 0, or just start the scale from 1. The following code should give you and idea of how to do either, or both. To remove 0s from a plot it might be easiest to just turn them into NAs, which won't be plotted (it will throw a warning, but that's okay). You can change your size scale breaks using scale_size_continuous:
Data2 %>%
mutate(frequency = ifelse(frequency == 0, NA, frequency),
year = as.factor(year)
) %>%
ggplot(aes(year, size = frequency, variable, color = variable)) +
geom_point() +
theme(text = element_text(size = 18)) +
scale_size_continuous(breaks = c(1, 4, 8, 12))

Answer 2

When the axis labels are not all showing up, or when they are superimposed, a standard trick is to rotate them. In this case I will rotate the x axis labels by 45 degrees.
library(ggplot2)
ggplot(Data2, aes(year, variable,
size = frequency, color = variable)) +
geom_point() +
theme(text = element_text(size = 18),
axis.text.x = element_text(angle = 45, hjust = 1))

Answer 3

I have a quick&dirty solution for no 0 frequencies: Data2[Data2==0] <- NA
alternatively only for frequency:
library(data.table)
setDT(Data2)
Data2[frequency==0, frequency:= NA]
For having proper years:
Data2$year = as.Date(strptime(Data2$year, "%Y"))
ggplot(Data2, aes(year, variable, size = frequency, color = variable)) +
geom_point()+
theme( axis.text.x = element_text(angle = 60, hjust = 1) )