CODEWITHSUNDEEP
rggplot2
RPo
RPo

Reputation: 35

How to overlay two plots with different variable treatment?

I am trying to overlay two plots with different variable treament (from the same dataset) : - the first one with a mutate and reorder (to make a geom_boxplot + geom_jitter) ; - the second one with a group_by and summarize (to make a geom_line). Both of them have to be overlayed.

When I try the following code, this Error is given : Aesthetics must be either length 1 or the same as the data (4): label

Local <- c("A", "B", "C", "D", "A", "B", "C", "D")
Case <- c("QQ", "DD", "GG", "PP", "QQ", "DD", "GG", "PP")
Div <- c(2, 4, 5, 1, 3, 5, 6, 7)
dat <- data.frame(Local, Case, Div)

p1 <- dat %>%
  mutate(Loc = reorder(Local, Div, FUN = median)) %>%
  ggplot(aes(Loc, Div, label = Case)) +
  geom_boxplot(outlier.size = -1) +
  geom_jitter(width = 0.1, alpha = 1, aes(color = Case, size = Div)) +
  geom_text_repel()

dat %>%
  group_by(Local) %>% 
  summarise(Div = mean(Div)) %>%
  mutate(Loc = reorder(Local, Div, FUN = median)) %>%
  ggplot(aes(Loc, Div, group = 1)) +
  geom_line()

p1 + geom_line (data = dat %>%
              group_by(Local) %>% 
              summarise(Div = mean(Div)) %>%
              mutate(Loc = reorder(Local, Div, FUN = median)), aes(Loc, Div, group = 1))

The first plot gives :

Plot1

And the second one :

Plot1

But how to overlay them ?

Upvotes: 0

Views: 172

Answers (1)

tom
tom

Reputation: 725

datBox <- dat %>%
  mutate(Loc = reorder(Local, Div, FUN = median)) 
datLine <- dat %>%
  group_by(Local) %>% 
  summarise(Div = mean(Div)) %>%
  mutate(Loc = reorder(Local, Div, FUN = median)) %>% 
  mutate(LocNum = recode(Loc, A = "1", D="2", B="3", C="4"))


ggplot(data = datBox, aes(Loc, Div)) +
geom_boxplot(outlier.size = -1) +
  geom_jitter(width = 0.1, alpha = 1, aes(color = Case, size = Div)) +
  geom_text_repel(aes(label = Case)) + 
  geom_line(data =datLine, aes(as.numeric(LocNum),Div))

I made a secondary axis with numeric values corresponding to the order of the Loc factor. geom_line does not appreciate the factor in axis. There is propably more elegant solutions. Also I inserted label in the geom_text_repel aes

Upvotes: 1

Related Questions