geom_text positioning labels individually

Question

I’m making a chart similar to this one in the example, where I position the labels by group: I’m using the code:

set.seed(1000)

ano=factor(rep(1996:2016,2))
cod=c(rep("K1",21),rep("O1.3",21))
valor=round(c(cumsum(rnorm(21)),cumsum(cumsum(rnorm(21)))),1)

dat=data.frame(cod,ano,valor)

ggplot(dat,aes(x=ano,y=valor, group=cod, color=cod, label=valor)) +
  geom_line(aes(linetype = cod), size=0.5) +
  geom_point(aes(shape=cod), size=2.5) +
  geom_text(data=subset(dat,cod=="K1"), aes(y = valor + 2), size=3) + 
  geom_text(data=subset(dat,cod=="O1.3"), aes(y = valor - 2), size=3) +
  scale_colour_manual(values = c("red", "darkblue")) +
  theme_minimal() +
  theme(legend.position = "bottom",
        legend.title = element_blank(),
        axis.title.x = element_blank(),
        axis.text.y  = element_blank(),
        axis.title.y = element_blank())

The result I would like is for the position of the labels to accompany the respective points, that is, when the blue line is above the red line, the corresponding blue labels are close to the respective points, above them, and vice versa. The same for the red line. Can anyone help me?

Answer 1

What you need to do is use the argument nudge_y of geom_text. You must pass a vector of values of the same length as the original. Follow code:

# Cria as posições
dat$pos <- c(ifelse(dat[dat$cod=="K1", 3] > dat[dat$cod=="O1.3", 3], 1, -1),
  ifelse(dat[dat$cod=="O1.3", 3] > dat[dat$cod=="K1", 3], 1, -1))

ggplot(dat,aes(x=ano,y=valor, group=cod, color=cod, label=valor)) +
  geom_line(aes(linetype = cod), size=0.5) +
  geom_point(aes(shape=cod), size=2.5) +
  geom_text(data=dat, aes(y = valor), size=3, nudge_y = dat$pos) + 
  scale_colour_manual(values = c("red", "darkblue")) +
  theme_minimal() +
  theme(legend.position = "bottom",
        legend.title = element_blank(),
        axis.title.x = element_blank(),
        axis.text.y  = element_blank(),
        axis.title.y = element_blank())

In cases of overlap (for different values of the set.seed(), just change the rule.

Answer 2

Try the following. The trick is to find a multiplier for the distance to add to the position y in the geom_text, in this case 2. I’ll do it with ave.

mult <- ave(dat$valor, dat$ano, FUN = function(x) 2*(x[1] < x[2]) - 1)
mult <- mult*(2*(dat$cod != "K1") - 1)

ggplot(dat,aes(x=ano,y=valor, group=cod, color=cod, label=valor)) +
  geom_line(aes(linetype = cod), size=0.5) +
  geom_point(aes(shape=cod), size=2.5) +
  geom_text(aes(y = valor + mult*2), size=3) + 
  scale_colour_manual(values = c("red", "darkblue")) +
  theme_minimal() +
  theme(legend.position = "bottom",
        legend.title = element_blank(),
        axis.title.x = element_blank(),
        axis.text.y  = element_blank(),
        axis.title.y = element_blank())

Answer 3

Using together the excellent solutions received I arrived at the following code:

mult = 2*(dat[dat$cod==rots[1], 3] > dat[dat$cod==rots[2], 3])-1
dat$pos=c(mult,mult*(-1))

ggplot(dat,aes(x=ano,y=valor, group=cod, color=cod, label=valor)) +
  geom_line(aes(linetype = cod), size=0.5) +
  geom_point(aes(shape=cod), size=2.5) +
  geom_text(data=dat, aes(y = valor + dat$pos*desloc), size=3,
            show.legend = FALSE) + 
  scale_colour_manual(values = c("red", "darkblue")) +
  theme_minimal() +
  theme(legend.position = "bottom",
        legend.title = element_blank(),
        axis.title.x = element_blank(),
        axis.text.y  = element_blank(),
        axis.text.x = element_text(angle = 90),
        axis.title.y = 
          element_text(margin = margin(t = 0, r = 20, b = 0, l = 0)))