I want to plot with ggplot the frequency of values from a numeric vector. With plot() is quite straight forward but I can't get the same result with ggplot.
library(ggplot2)
dice_results <- c(1,3,2,4,5,6,5,3,2,1,6,2,6,5,6,4)
hist(dice_results)
ggplot(dice_results) + geom_bar()
# Error: ggplot2 doesn't know how to deal with data of class numeric
Should I create a dataframe for ggplot() to plot my vector?
Please look at the help page ?geom_histogram. From the first example you may find that this works.
qplot(as.factor(dice_results), geom="histogram")
Also look at ?ggplot. You will find that the data has to be a data.frame
Try the code below
library(ggplot2)
dice_results <- c(1,3,2,4,5,6,5,3,2,1,6,2,6,5,6,4,1,3,2,4,6,4,1,6,3,2,4,3,4,5,6,7,1)
ggplot() + aes(dice_results)+ geom_histogram(binwidth=1, colour="black", fill="white")
The reason you got an error - is the wrong argument name. If you don't provide argument name explicitly, sequential rule is used - data arg is used for input vector.
To correct it - use arg name explicitly:
ggplot(mapping = aes(dice_results)) + geom_bar()
You may use it inside geom_ functions familiy without explicit naming mapping argument since mapping is the first argument unlike in ggplot function case where data is the first function argument.
ggplot() + geom_bar(aes(dice_results))
Use geom_histogram instead of geom_bar for Histogram plots:
ggplot() + geom_histogram(aes(dice_results))
Don't forget to use bins = 5 to override default 30 which is not suitable for current case:
ggplot() + geom_histogram(aes(dice_results), bins = 5)
qplot(dice_results, bins = 5) # `qplot` analog for short
To reproduce base hist plotting logic use breaks args instead to force integer (natural) numbers usage for breaks values:
qplot(dice_results, breaks = 1:6)
来源：https://stackoverflow.com/questions/19025717/how-to-plot-histogram-frequency-count-of-a-vector-with-ggplot