Trying to make a list of ggplot objects in a for loop; all items in list are written as last iteration from loop

Question

I want to build a list of ggplot objects. I want to plot one variable on the x axis but plot each of the other variables in the data frame in the y. When I run the loop all of my values end up being the same as the last variable.

I'm hoping to use the grid.arrange function to plot all the graphs on one page.

My reprex:

library(ggplot2)

 l <- list()
for(i in 2:11){
    p <- ggplot(mtcars, aes(x = mpg, y = mtcars[,i])) + geom_smooth() + geom_point()
    name <- paste("p",i,sep="_")
    tmp <- list(p)
    l[[name]] <- tmp
}
print(l$p_2)
print(l$p_3)

Using a loop probably isn't the best way to solve the issue. It's better to describe your original problem rather than your assumed solution to the problem. Would something like this help? https://stackoverflow.com/questions/4877357/how-to-plot-all-the-columns-of-a-data-frame-in-r — Jack Brookes, Mar 08 '18 at 21:57
I did describe my original problem in the first 2 sentences. The link you sent is not the same. I cannot plot on the same graph because my variables have unique scales like they do in the reprex. I looked at facet_grid but it uses a static value for aes(y =). — dhbrand, Mar 08 '18 at 22:07
Here's my tidyverse solution which I think is a lot cleaner than the ones below, now the question is closed https://gist.github.com/jackbrookes/3bbcc47cdd246b3d8a9173a4b9bacf43 — Jack Brookes, Mar 08 '18 at 22:23

eipi10 · Accepted Answer · 2018-03-08T22:16:43.427

You can create a list of plots directly using sapply. For example:

plist = sapply(names(mtcars)[-grep("mpg", names(mtcars))], function(col) {
  ggplot(mtcars, aes_string(x = "mpg", y = col)) + geom_smooth() + geom_point()
}, simplify=FALSE)

The list elements (each of which is a ggplot object) will be named after the y variable in the plot:

names(plist)

[1] "cyl"  "disp" "hp"   "drat" "wt"   "qsec" "vs"   "am"   "gear" "carb"

You can print all the plots by typing plist in the console. Or, for a single plot, just select the plot you want:

plist[["hp"]]

For a situation like this, you might prefer faceting, which requires converting the data from wide to long format. You can have facets with different y scales by setting scales="free_y".

library(tidyverse)

ggplot(gather(mtcars, key, value, -mpg), aes(mpg, value)) + 
  geom_smooth() + geom_point() +
  facet_wrap(~ key, scales="free_y", ncol=5)

www · Answer 2 · 2018-03-08T22:10:08.253

2

Subset the data frame before calling ggplot, and then use aes_string to call column names as string.

library(ggplot2)

l <- list()
for(i in 2:11){
  temp <- mtcars[, c(1, i)]
  p <- ggplot(temp, aes_string(x = "mpg", y = names(temp)[2])) + geom_smooth() + geom_point()
  name <- paste("p",i,sep="_")
  tmp <- list(p)
  l[[name]] <- tmp
}
print(l$p_2)

print(l$p_3)

edited Mar 08 '18 at 22:10

answered Mar 08 '18 at 22:03

www

38,575
12
48
84

Trying to make a list of ggplot objects in a for loop; all items in list are written as last iteration from loop

2 Answers2

Linked

Related