雷达图/带有ggplot2的蜘蛛图

Question

I am trying to make a radar chart with ggplot2.

> Test1
   Country fertility gengap   LEB
1 Colombia      1.92   0.29 79.30
2     Peru      1.94   0.37 78.40
3      Min      9.23   3.83  0.00
4      Max      1.59  23.70 78.43

#normalizing the data:

Test1_norm<- data.frame(sapply(Test1[1:4, 2:4], scales::rescale))

    > Test1_norm
   fertility      gengap       LEB
1 0.04319372 0.000000000 1.0000000
2 0.04581152 0.003417343 0.9886507
3 1.00000000 0.151217428 0.0000000
4 0.00000000 1.000000000 0.9890290

I found this code to plot a radar chart: http://www.cmap.polytechnique.fr/~lepennec/R/Radar/RadarAndParallelPlots.html for the mtcars-example:

   ggplot(mtcarsmelted, aes(x = variable, y = value)) +
  geom_path(aes(group = model, color = model),  size = 2) +
  theme(strip.text.x = element_text(size = rel(0.8)),
        axis.text.x = element_text(size = rel(0.8)),
        axis.ticks.y = element_blank(),
        axis.text.y = element_blank()) +
  xlab("") + ylab("") +
  guides(color = guide_legend(ncol=2)) +
  coord_polar()

Using the mtcars example everything works fine for me, but when I try to rewrite the code for my data I just get errors. I was trying for several hours now and I am not sure which variables I need to exchange and which have to stay like in the example.

I would appreciate a little help a lot! I am still very new to R ;)

Thanks everybody :)

Answer 1

First, using normal data, I attempt to plot radar charts according to the method you display to me. reshape2 package is what you need. Here's the code:

Test1_norm <- data.frame(group=c("Colombia","Peru","Min","Max"),
              fertility=c(0.04319372,0.04581152,1,0),
              gengap=c(0,0.003417343,0.151217428,1),
              LEB=c(1,0.9886507,0,0.9890290))
Test1_norm$model <- Test1_norm$group
library(reshape2)
Test1_plot<- melt(Test1_norm)

##Define a radar coordinate system
coord_radar <- function (theta = "x", start = 0, direction = 1) 
{
    theta <- match.arg(theta, c("x", "y"))
    r <- if (theta == "x") 
        "y"
    else "x"
    ggproto("CordRadar", CoordPolar, theta = theta, r = r, start = start, 
        direction = sign(direction),
        is_linear = function(coord) TRUE)
}

#Plot
library(ggplot2)
ggplot(Test1_plot, aes(x = variable, y = value)) +
geom_polygon(aes(group = model, color = model), fill = NA, size = 2, show.legend = FALSE) +
  geom_line(aes(group = model, color = model), size = 2) +
  theme(strip.text.x = element_text(size = rel(0.8)),
        axis.text.x = element_text(size = rel(0.8)),
        axis.ticks.y = element_blank(),
        axis.text.y = element_blank()) +
  xlab("") + ylab("") +
  guides(color = guide_legend(ncol=2)) +
  coord_radar()

Fortunately, Ricardo Bion provided an efficient methods to create radar charts with ggplot2.This may be helpful to you: https://github.com/ricardo-bion/ggradar You can install the package and call ggradar() function as introduced in the manual.It's my code with original data.

Test1 <- data.frame(fertility=c(1.92,1.94,9.23,1.59),
                    gengap=c(0.29,0.37,3.83,23.70),
                     LEB=c(79.30,78.40,0.00,78.43))
rownames(Test1)<-c("Colombia","Peru","Min","Max")

source("ggradar.R")
library(dplyr)
library(ggplot2)
library(scales)
Test1 %>%
      add_rownames( var = "group" ) %>%
      mutate_each(funs(rescale), -group)  -> Test1_radar
ggradar( Test1_radar)
ggsave("Test1_radar.tiff",width = 8, height = 8)