Question

我正在尝试向我的ggplot添加错误条带，但由于某种原因，阴影与绘制的线条不匹配。这是我第二次遇到这个问题，我无法弄清楚为什么会发生这种情况。最初，我认为这可能是由于某些变量被定义为因素的方式，但即使我更改其格式，我仍然会遇到此问题。任何想法或建议将不胜感激！谢谢！

Plot with incorrect ribbons

数据框：

library(ggplot2)
limits <- aes(ymax = m$Mean + m$SE, ymin=m$Mean - m$SE)

Dplot<- ggplot(data= m, aes(x=sound, y= Mean, color= `Reading condition`, 
                            fill= `Reading condition`,
                            group=`Reading condition`, shape=`Reading condition`,
                            linetype=`Reading condition`))+ 
  scale_fill_brewer(palette="Dark2")+ scale_colour_brewer(palette="Dark2")+
  theme_bw() + theme(panel.grid.major = element_line(colour = "#E3E5E6", size=0.7), 
                     axis.line = element_line(colour = "black", size=1),
                     panel.border = element_rect(colour = "black", size=1, fill = NA))+
  geom_line(size=2) + ylim(0, 0.5)+
  geom_point(size=7)+ scale_x_discrete(expand=c(0.2,0.2))+
  #scale_y_continuous(limits- c(0,0.50))+
  xlab("\n Background sound")+ ylab("Comprehension above chance level (%)")+ 
  theme(legend.position="bottom", legend.title=element_text(size=20, face="bold", family="serif"), legend.text=element_text(size=20,family="serif"),legend.key.width=unit(2,"cm"),
        legend.key.height=unit(1,"cm"), strip.text=element_text(size=20, family="serif"),
        title=element_text(size=20, family="serif"),
        axis.title.x = element_text(size=20, face="bold", family="serif"), axis.title.y = element_text(size=20, face="bold", family="serif"), 
        axis.text=element_text(size=20, family="serif"), 
        panel.border = element_rect(linetype = "solid", colour = "black"), 
        legend.key = element_rect(colour = "#000000", size=1))+
  geom_ribbon(limits, alpha=0.15, colour=NA) +
  facet_grid(.~ diff) + theme(strip.text.x = element_text(size = 20,  face="bold",family="serif"),
                                 strip.background = element_rect(fill="#F5F7F7", colour="black", size=1.5),
                                 legend.key = element_rect(colour = "#000000", size=1))

图表代码：

{{1}}

Answer 1

指定数据框并对变量进行子集化（例如使用$）可能会导致ggplot中发生奇怪的事情。您永远不需要这样的子集，因为数据帧已经在某个时刻指定给data参数（可能在第一个ggplot(...)调用中）。数据，如美学，也从ggplot继承到geoms和stats，所以除非你改变数据，否则没有理由担心重新指定它的来源。（如果您执行想要更改使用的数据框，请指定相关geom或stat的data参数。）

删除m$将解决问题。重新排列和简化使逻辑更加明显，

library(ggplot2)

m <- data.frame(sound = c("English", "English", "English", "English", "Silence", "Silence", "Silence", "Silence"), 
                `Reading condition` = c("normal", "normal", "trailing mask", "trailing mask", "normal", "normal", "trailing mask", "trailing mask"), 
                diff = c("Difficult questions", "Easy questions", "Difficult questions", "Easy questions", "Difficult questions", "Easy questions", "Difficult questions", "Easy questions"), 
                Mean = c(0.283, 0.36, 0.183, 0.227, 0.19, 0.347, 0.197, 0.333), 
                SD = c(0.500558971667007, 0.34814947033031, 0.497195715406262, 0.447163568809774, 0.49804988833968, 0.361515576848703, 0.498812908487118, 0.373926502247132), 
                SE = c(0.100111794333401, 0.0696298940660619, 0.0994391430812524, 0.0894327137619548, 0.099609977667936, 0.0723031153697406, 0.0997625816974236, 0.0747853004494263), 
                check.names = FALSE)

ggplot(m, aes(x = sound, y = Mean, ymax = Mean + SE, ymin = Mean - SE, 
              color = `Reading condition`, fill = `Reading condition`, 
              group = `Reading condition`)) + 
    geom_line() +
    geom_point() + 
    geom_ribbon(alpha = 0.15, colour = NA) +
    facet_grid(. ~ diff)

Answer 2

通过这种方式考虑Alistaire的正确答案：ggplot采用您的起始数据框架，并应用您的ggplot代码所需的所有变换，比例和统计数据。然后在结果坐标系中生成图形。

通过将功能区的aes锚定到具有m $ ...的全局环境data.frame，您将指定预变换，pre-stat，预缩放坐标...有时您的原始坐标系可能是容易混淆地靠近隐藏的坐标系，以便你的元素出现，但只是有点偏离，正如你所看到的。

ggplot2：标准错误色带与绘图线不匹配

2 个答案: