CODEWITHSUNDEEP
rggplot2
Dan P
Dan P

Reputation: 43

geom_vline and facet_wrap error in ggplot

I am trying to run the following code:

temp_plotdata <- data.table(Treatment_Code = c('Control', 'Control', 
                                               'Second Mailing', 'Second Mailing', 
                                               'First Mailing', 'First Mailing'),
                            Q9 = c(9, 14, 10, 3, 1, 4))

output_gg <-
  ggplot(temp_plotdata, aes(x = Q9)) +
  geom_histogram(binwidth = 1, fill = 'lightblue') +
  geom_vline(data = temp_plotdata[, summary(Q9)[c(2,3,5)], by=Treatment_Code],
             aes(xintercept = V1),
             linetype = 'dashed', color = 'darkred') +
  facet_wrap(~Treatment_Code, ncol = 1)

I am getting back an error of:

Error in provideDimnames(x, sep = sep, base = base) : 'dimnames' applied to non-array

I know the problem is in the geom_vline part of the code because when I run it without those lines or run it with something like geom_vline(xintercept = c(3, 5, 8)) it works fine. I also tried turning the data from the geom_vline into a separate dataframe first, but it did not work.

I ran a very similar piece of code last year and it worked fine, so I'm not sure if something changed with geom_vline or if my code is just not correct due to new data or some small change I might have accidentally made.

Thanks for any help you can give.

Upvotes: 4

Views: 294

Answers (1)

eipi10
eipi10

Reputation: 93811

This is happening because the class of V1 (the summary column returned by data.table) is table, rather than a numeric vector. Change it to a vector and it should work.

output_gg <-
  ggplot(temp_plotdata, aes(x=Q9)) +
  geom_histogram(binwidth=1, fill='lightblue') +
  geom_vline(data=temp_plotdata[, as.vector(summary(Q9)[c(2,3,5)]), by=Treatment_Code],
             aes(xintercept=V1),
             linetype='dashed', color='darkred') +
  facet_wrap(~ Treatment_Code, ncol=1)

Compare the structure of the data frame before and after:

str(temp_plotdata[, summary(Q9)[c(2,3,5)], by=Treatment_Code])

Classes ‘data.table’ and 'data.frame':    9 obs. of  2 variables:
  $ Treatment_Code: chr  "Control" "Control" "Control" "Second Mailing" ...
  $ V1            :Class 'table'  num [1:9] 10.25 11.5 12.75 4.75 6.5 ...
- attr(*, ".internal.selfref")=<externalptr>

str(temp_plotdata[, as.vector(summary(Q9)[c(2,3,5)]), by=Treatment_Code])

Classes ‘data.table’ and 'data.frame':    9 obs. of  2 variables:
  $ Treatment_Code: chr  "Control" "Control" "Control" "Second Mailing" ...
  $ V1            : num  10.25 11.5 12.75 4.75 6.5 ...
- attr(*, ".internal.selfref")=<externalptr>

Upvotes: 2

Related Questions