newbie in R: wanted to to ask how I can plot two numerical variables (say smoking% and liquor consumption%) across 50 US states - I can easily do that in excel but still struggling in R; any ideas?enter image description here
#Let's say here is my code:
smoking <- c(21, 23.4, 83)
drinking <- c(19.5, 28.9, 57)
States <- c(CA, NH, NJ)
How can i plot as barplot for smoking (red) and drinking (black) with states on x axis. Sorry Stacko won't allow me to upload pic or data file. Thanks again!
The data you provided, when it is put in a table, is not 'tidy'. You can read more about tidy data here.
Here is what I have done to get your desired output, assuming you start with data like df:
smoking <- c(21, 23.4, 83)
drinking <- c(19.5, 28.9, 57)
States <- c("CA", "NH", "NJ")
# the 'messy' data
df <- data.frame("States" = States,
"Smoking" = smoking,
"Drinking" = drinking)
> df
States Smoking Drinking
1 CA 21.0 19.5
2 NH 23.4 28.9
3 NJ 83.0 57.0
When can then use the tidyr package to transform it into long format. This is much easier to work with in ggplot2.
library(tidyr)
df_tidy <- pivot_longer(df, -States, names_to = "Type")
> df_tidy
# A tibble: 6 x 3
States Type value
<chr> <chr> <dbl>
1 CA Smoking 21
2 CA Drinking 19.5
3 NH Smoking 23.4
4 NH Drinking 28.9
5 NJ Smoking 83
6 NJ Drinking 57
Finally, we can plot this using ggplot2:
library(ggplot2)
ggplot(df_tidy, aes(x = States, y = value)) +
geom_bar(aes(fill = Type), position = "dodge", stat = "identity")
We can also just use geom_col() instead of geom_bar(), which uses stat = "identity" by default, as we find out from this tidyverse help page.
ggplot(df_tidy, aes(x = States, y = value)) +
geom_col(aes(fill = Type), position = "dodge")
In either case, we get:
