display a secondary x axis category

[freddywit]

Ciao guys,

I have the following dataframe

obj <- data.frame (percentile = c(0.1, 0.2, 0.3, 0.4, 0.5, 0.6 , 0.7, 0.8, 0.9, 1),
                   emp_change = c(0.05, 0.04, 0.03, 0.05, 0.06, 0.04, 0.02, 0.09, 0.08, 0.06), 
                   task = c("Manual", "Manual", "Manual", "Routine-Manual", "Routine-Manual",
                            "Routine-Abstract", "Routine-Abstract", "Abstract", "Abstract", "Abstract"))

My goal is to display the variable task on my x axis as well. I don't know if it would be necessary to generate a second x axis. I know that the question is pretty straightforward and was definitely asked before but I fail as I only found solutions for barplots.

This is my plot. Note also, that I use in my real data geom_smooth rather than geom_line. But I guess this should be irrespective when creating a second x axis.

plot <- ggplot() +
  geom_line(data = obj, aes(x= percentile, y= emp_change, group = 1, color="observed", linetype = "observed"), 
      size=1.5, linetype = "dotdash")
print(plot +  theme(axis.text.x=element_text(angle = 60, hjust = 1)) + 
        theme(axis.title=element_text(size=12)) +
        labs(y="100 x Change in Employment Share", x = "Percentile ranked by task input and log of mean occupational wage (1990)"))

Many thanks in advance

Freddy

[FJCC]

I suggest you try a couple of things other than a second x axis: colored points with the color encoding the task or geom_text to label the points. I have rough versions of each below. For the colored points, I would probably also change the line to a simple line instead of dashed. My geom_text version is rather ugly but you could adjust the text placement, even making y a constant value.

library(ggplot2)
obj <- data.frame (percentile = c(0.1, 0.2, 0.3, 0.4, 0.5, 0.6 , 0.7, 0.8, 0.9, 1),
                   emp_change = c(0.05, 0.04, 0.03, 0.05, 0.06, 0.04, 0.02, 0.09, 0.08, 0.06), 
                   task = c("Manual", "Manual", "Manual", "Routine-Manual", "Routine-Manual",
                            "Routine-Abstract", "Routine-Abstract", "Abstract", "Abstract", "Abstract"))

#colored points
plot <- ggplot() +
  geom_line(data = obj, 
            aes(x= percentile, y= emp_change, group = 1, ), 
            size=1.5, linetype = "dotdash")+
  geom_point(data = obj, 
            aes(x= percentile, y= emp_change,color=task),size=2)
print(plot +  theme(axis.text.x=element_text(angle = 60, hjust = 1)) + 
        theme(axis.title=element_text(size=12)) +
        labs(y="100 x Change in Employment Share", x = "Percentile ranked by task input and log of mean occupational wage (1990)"))

#geo_text
plot <- ggplot() +
  geom_line(data = obj, 
            aes(x= percentile, y= emp_change, group = 1, ), 
            size=1.5)+
  geom_text(data = obj, 
             aes(x= percentile, y= emp_change+0.005,label=task),vjust=0)
print(plot +  theme(axis.text.x=element_text(angle = 60, hjust = 1)) + 
        theme(axis.title=element_text(size=12)) +
        labs(y="100 x Change in Employment Share", x = "Percentile ranked by task input and log of mean occupational wage (1990)"))

^{Created on 2021-05-15 by the reprex package (v0.3.0)}

[freddywit]

thanks for your response! But the problem is that my real data is based on geom_smooth, so a local regression. In addition, there are 300 observations, for which reason the ideas above would not be as descriptive.
Thanks for your help anyways