如何通过在 R studio 中使用 ggplot 在条形图中组合 2 个变量？

Question

I'm trying to plot a multivariate bar chart by using ggplot such that one bar consist of two different colour which represent two variables instead of comparing the 2 variables side by side. The 2 variables will be positive probability ( P.Probability ) and negative probability ( N.Probability ).

The code below is my data. I'm only able to include P.Probability in the code but I want both P.Probability and N.Probability to be included in the graph as I mentioned above.

Month <- c("Aug", "Sep", "Oct")
P.Probability <- c(0.5, 0.6, 0.6)
N.Probability <- 1-P.Probability

dtf2 <- data.frame(Month, P.Probability, N.Probability)

ggplot(dtf2, aes(x = Month, y = P.Probability)) + 
  geom_bar(stat = "identity") + 
  geom_hline(yintercept=0)+
  theme(axis.title.y=element_blank(),
        axis.title.x=element_blank(),
        legend.position = "none",
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5))+
  theme(panel.background = element_blank(),
        axis.ticks.x = element_blank()) +
  theme(axis.line.y = element_line(color="black", size = 0.5))+
  geom_text(aes(label = paste(P.Probability*100, "%"), vjust = ifelse(P.Probability >= 0, -0.5, 1.2)))+
  scale_y_continuous(labels=scales::percent)

The diagram attached below is the graph that I want. I hope the N.Probability will be the blue colour as the bottom and the P.Probability will be the red colour on top.

Answer 1

I'd suggest gathering into long format so you can map the type of probability to fill color.

dtf2_long <- tidyr::gather(dtf2, type, Probability, -Month)

ggplot(dtf2_long, aes(x = Month, y = Probability, fill = type)) + 
  geom_bar(stat = "identity") + 
  geom_hline(yintercept=0)+
  theme(axis.title.y=element_blank(),
        axis.title.x=element_blank(),
        legend.position = "none",
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5))+
  theme(panel.background = element_blank(),
        axis.ticks.x = element_blank()) +
  theme(axis.line.y = element_line(color="black", size = 0.5))+
  geom_text(data = dtf2_long %>% filter(type == "P.Probability"),
            aes(label = paste(Probability*100, "%"), vjust = ifelse(Probability >= 0, -0.5, 1.2)))+
  scale_y_continuous(labels=scales::percent)