ggplot2，註釋y軸以指示所用刻度的方向

Question

我製作了一條反映感知亮度失真的（連續）y軸的條形圖。我的數據被測量的方式是，零以上的數字表示這些對象被判斷爲比實際情況要輕，低於零的數字表示這些對象被判斷爲比實際情況更暗。我怎樣才能在y軸旁放置一些文字來指示我使用的刻度的方向，即「更深 - 更輕」？我仍然希望將當前的y軸標籤保持爲「亮度失真（灰度級）」，但只需向讀者添加一個說明比例的註釋。

這裏是我的數據樣本：

"Subject" "Cond" "Distortion" 
"1" "White" 10.7 
"2" "White" 19.4 
"3" "White" 15.9 
"4" "White" 13.5 
"5" "White" 15.4 
"1" "Ambiguous" 13.4 
"2" "Ambiguous" 11.4 
"3" "Ambiguous" 8.9 
"4" "Ambiguous" 11.0 
"5" "Ambiguous" 11.4 
"1" "Black" 8.4 
"2" "Black" 1.7 
"3" "Black" 7.7 
"4" "Black" 8.0 
"5" "Black" 5.7 

，併產生我的條形圖代碼：

library(ggplot2) 
g <- ggplot(data, aes(x = Cond, y = Distortion)) 
g + stat_summary(fun.y = mean, geom="bar",position="dodge", 
       fill = "Gray",colour="Black") + 
     stat_summary(fun.data=mean_cl_normal, geom="errorbar", 
        position=position_dodge(width=0.90), width=0.2) + 
     labs(x="Faces", y = "Lightness Distortion (gray levels)")+ 
     theme_bw() 

sample graph

感謝您的閱讀！

Answer 1

我不太確定你以後的樣子，但是這顯示了兩種方法在文字面板外定位文字。對於這兩種方法，第一項任務是創建一個新的文本grob，其中包含文本「更輕」和「更深」。第一種方法使用gtable函數定位新grob，gtable_add_grob;第二種方法使用ggplot函數定位grob，annotation_custom。

# Your data 
data = read.table(text = ' "Subject" "Cond" "Distortion" 
"1" "White" 10.7 
"2" "White" 19.4 
"3" "White" 15.9 
"4" "White" 13.5 
"5" "White" 15.4 
"1" "Ambiguous" 13.4 
"2" "Ambiguous" 11.4 
"3" "Ambiguous" 8.9 
"4" "Ambiguous" 11.0 
"5" "Ambiguous" 11.4 
"1" "Black" 8.4 
"2" "Black" 1.7 
"3" "Black" 7.7 
"4" "Black" 8.0 
"5" "Black" 5.7', header = TRUE) 

library(ggplot2) 
library(gtable) 
library(grid) 

# The plot 
p <- ggplot(data, aes(x = Cond, y = Distortion)) + 
    stat_summary(fun.y = mean, geom = "bar", position = "dodge", 
     fill = "Gray", colour = "black") + 
    stat_summary(fun.data = mean_cl_normal, geom = "errorbar", 
     position = position_dodge(width = 0.90), width = 0.2) + 
    labs(x = "Faces", y = "Lightness Distortion (gray levels)") + 
    theme_bw() 

## Using `gtable` to position the text grob 
# Set up text grob 
text.grob1 <- textGrob("Darker", y = 0, just = "left", 
    rot = 90, gp = gpar(fontsize = 10, col = "red", fontface = "italic")) 

text.grob2 <- textGrob("Lighter", y = 1, just = "right", 
    rot = 90, gp = gpar(fontsize = 10, col = "red", fontface = "italic")) 

# Combine them into a single grob 
text.grob <- gTree(children = gList(text.grob1, text.grob2)) 

# Get ggplot grob 
g <- ggplotGrob(p) 

# Add the grob to the column containing the y axis title 
pos = subset(g$layout, grepl("ylab-l", name), t:l) 
g <- gtable_add_grob(g, text.grob, l = pos$l, t = pos$t) 

# Draw it 
grid.newpage() 
grid.draw(g) 

## Using `annotation_custom` to position the text grob 
# Set up text grob 
text.grob1 <- textGrob("Lighter", y = 0, just = "left", 
    rot = 90, gp = gpar(fontsize = 10, col = "red", fontface = "italic"), 
    vjust = -2.5) 

text.grob2 <- textGrob("Darker", y = 1, just = "right", 
    rot = 90, gp = gpar(fontsize = 10, col = "red", fontface = "italic"), 
    vjust = -2.5) 

# Combine them into a single grob 
text.grob <- gTree(children = gList(text.grob1, text.grob2)) 

# Add the grob to the plot 
p <- p + annotation_custom(text.grob, 
     xmin = -Inf, xmax = -Inf, ymin = Inf, ymax = -Inf) 

# The new grob is positioned outside the plot panel. 
# To see the grob, clipping needs to be turned off. 
g = ggplotGrob(p) 
g$layout$clip <- "off" 

# Draw it 
grid.newpage() 
grid.draw(g)