How do I drop some categorical variables from a plot in R?-CodePudding

So, I have this scatterplot:

data(infmort, package = "faraway")
summary(infmort)
#install.packages("ggplot2")
library(ggplot2)
#levels(infmort$region)

# Levels are : 1) Africa, 2) Europe, 3) Asia, 4), Americas

plot(mortality ~ log(income), data = infmort, col = region_col, pch = region_col)
legend("topleft", legend = c("Africa", "Europe", "Asia", "Americas"), col = 1:4, pch = 1:4)

I want to drop the levels of "Africa" and "Europe", so that my scatterplot is only showing the data for Asia and Americas. How do I go about doing this? I am new to R.

Here is what I tried to drop these two categorial variables:

plot(mortality ~ log(income), data=infmort, col = typ_col, pch = typ_col)
legend("topleft", legend = c("Americas", "Asia"), col = 1:2, pch= 1:2)

I know that this is incorrect because it is still including all of the data, I was just wondering how I only graph the data for Americas and Asia.

CodePudding user response：

One potential solution is to subset the dataframe before plotting, e.g.

#install.packages("faraway")
data(infmort, package = "faraway")
summary(infmort)
#>       region       income         mortality                  oil    
#>  Africa  :34   Min.   :  50.0   Min.   :  9.60   oil exports   : 9  
#>  Europe  :18   1st Qu.: 123.0   1st Qu.: 26.20   no oil exports:96  
#>  Asia    :30   Median : 334.0   Median : 60.60                      
#>  Americas:23   Mean   : 998.1   Mean   : 89.05                      
#>                3rd Qu.:1191.0   3rd Qu.:129.40                      
#>                Max.   :5596.0   Max.   :650.00                      
#>                                 NA's   :4
#install.packages("ggplot2")
library(ggplot2)
#levels(infmort$region)
# Levels are : 1) Africa, 2) Europe, 3) Asia, 4), Americas

region_col <- 1:4
plot(mortality ~ log(income), data = infmort, col = region_col, pch = region_col)
legend("topleft", legend = c("Africa", "Europe", "Asia", "Americas"), col = 1:4, pch = 1:4)

typ_col <- 1:2
plot(mortality ~ log(income),
     data=infmort[infmort$region %in% c("Americas", "Asia"),],
     col = typ_col, pch = typ_col)
legend("topleft", legend = c("Americas", "Asia"), col = 1:2, pch= 1:2)

^{Created on 2022-03-30 by the reprex package (v2.0.1)}

CodePudding user response：

You can simply filter the data prior to making a plot.

library(dplyr)
infmort <- infmort %>% dplyr::filter(region %in% c("Asia", "America"))