Create graphs by group using ggplot in R

Question

I'm relatively new to using ggplot2 in R and have been struggling with this for awhile. I have figured out how to get everything from one data frame on a graph (that is pretty easy...), and how to write a loop function to get each observation (id in the example below) onto their own graphs but not how to create separate graphs with multiple id per group, when the id and group can change each time I run the code. Here is some sample data and the output I am trying to produce.

x <- c(1,3,6,12,24,48,72,1,3,6,12,24,48,72,1,3,6,12,24,48,72,1,3,6,12,24,48,72)
y <- c(8,27,67,193,271,294,300,10,30,70,195,280,300,310,5,25,60,185,250,275,300,15,40,80,225,275,325,330)
group <- c(1,1,1,1,1,1,1,1,1,1,1,1,1,1,2,2,2,2,2,2,2,2,2,2,2,2,2,2)
id <- c(100,100,100,100,100,100,100,101,101,101,101,101,101,101,102,102,102,102,102,102,102,103,103,103,103,103,103,103)
df <- data.frame(x,y,group,id)

Similar questions were asked here and here but I still can't figure out how to do what I need because I need separate graphs (not facets) by group with multiple id on the same graph.

Edit to add attempt -

l <- unique(df$group)
for(l in df$group){
  print(ggplot(df, aes(x = x, y = y, group = group, color = id))+
    geom_line())
}

alexwhitworth · Answer 1 · 2022-01-21T19:29:36.550

1

Use facet_grid() or facet_wrap()

library(ggplot2)
ggplot(df, aes(x= x, y=y, colour= factor(id))) + geom_line() + facet_grid(group ~ .)

Edit: OP clarifies in comments they want separate graphs, not faceting

# 1
ggplot2(df[df$group == 1,], aes(x= x, y=y, colour= factor(id))) + geom_line()
# 2
ggplot2(df[df$group == 2,], aes(x= x, y=y, colour= factor(id))) + geom_line()

edited Jan 21 '22 at 19:29

answered Jan 21 '22 at 19:25

alexwhitworth

4,839
5
32
59

Thank you but I need separate graphs not facets – yaynikkiprograms Jan 21 '22 at 19:26
Then subset `df` and do two graphs – alexwhitworth Jan 21 '22 at 19:27
I know how to do that if I know what `id` are but I am trying to write a loop so I can use this for a routine weekly report – yaynikkiprograms Jan 21 '22 at 19:31
@yaynikkiprograms If your question doesn't reflect what you actually want to do, it's impossible to answer correctly. Kindly improve your question – alexwhitworth Jan 23 '22 at 00:58

score 1 · Accepted Answer · answered Jan 21 '22 at 22:44

To achieve your desired result

Split your dataframe by group using e.g. split
Use lapply to loop over the list of splitted data frames to create your plots or if you want to add the group labels to the title you could loop over names(df_split).

Note: I converted the id variable to factor. Also, you have to map id on the group aesthetic to get lines per group. However, as your x variable is a numeric there is actually no need for the group aesthetic.

library(ggplot2)

df_split <- split(df, df$group)

lapply(df_split, function(df) {
  ggplot(df, aes(x = x, y = y, group = id, color = factor(id))) +
    geom_line()
})

lapply(names(df_split), function(i) {
  ggplot(df_split[[i]], aes(x = x, y = y, group = id, color = factor(id))) +
    geom_line() +
    labs(title = paste("group =", i))
})
#> [[1]]

#> 
#> [[2]]

And even I if would recommend to use lapply the same could be achieved using a for loop like so:

for (i in names(df_split)) {
  print(
    ggplot(df_split[[i]], aes(x = x, y = y, group = id, color = factor(id))) +
      geom_line() +
      labs(title = paste("group =", i))
  )
}

Create graphs by group using ggplot in R

2 Answers2