ggplot2绘制了两个传说

时间:2013-03-12 19:33:20

标签: r ggplot2 legend legend-properties

我几乎完成了以下图表,但它有一个问题。

图表中的图例被绘制两次。

以下是数据:

structure(list(Period = c("January 1997 - August 2003", "September 2003 - Jun 2005", 
"Jul 2005 - Dec 2009", "January 1997 - August 2003", "September 2003 - Jun 2005", 
"Jul 2005 - Dec 2009"), Time.Period = structure(c(1L, 3L, 2L, 
 1L, 3L, 2L), .Label = c("Jan 1997 - Aug 2003", "Jul 2005 - Dec 2009", 
 "Sep 2003 - Jun 2005"), class = "factor"), Variable = structure(c(2L, 
 2L, 2L, 1L, 1L, 1L), .Label = c("Significant", "Zscore"), class = "factor"), 
 Score = c(8.798129, 4.267268, 7.280275, 1.64, 1.64, 1.64)), .Names = c("Period", 
 "Time.Period", "Variable", "Score"), class = "data.frame", row.names = c(NA, 
 -6L))

ggplot(glomor, aes(x=Time.Period, y=Score, group=Variable, shape=Variable, color=Variable)) +
 geom_point() +
 guides(fill=FALSE) +
 scale_x_discrete(limits=c("Jan 1997 - Aug 2003","Sep 2003 - Jun 2005","Jul 2005 - Dec 2009"),    expand=c(.08,0)) +
  geom_line(aes(linetype=Variable), size=1.5) +
  geom_point(size=4.2) +
  scale_linetype_manual(values=c(1,3)) +
  scale_color_manual(values=c("black", "grey40"), name="", labels=c("Signficant Z-Score",   "Moran's I Z-Score")) +
  scale_fill_discrete(name="", label=c("Signficant Z-Score", "Moran's I Z-Score")) +
  theme_classic()+
  ylim(0,10) +
  xlab("Time Periods") +
  ylab("Moran's I Z-Score") +
  theme(axis.title.x=element_text(size=14)) +
  theme(axis.title.y=element_text(size=14)) +
  theme(legend.position=c(.75, .85)) +
  theme(legend.background = element_rect(fill="white")) +
  theme(legend.key = element_blank())

有谁知道,为什么ggplot2产生两个传说?

1 个答案:

答案 0 :(得分:19)

您有三种映射到Variable的美学:形状,颜色和线型。当传说具有相同的标题和标签时,它们会折叠在一起。您已将标题设置为空白颜色,并为其指定自定义标签(“重要Z分数”和“Moran的I Z分数”)。你需要为线型和形状做到这一点,让它们一起崩溃。

更改

scale_linetype_manual(values=c(1,3)) +

scale_linetype_manual(values=c(1,3), name="", labels=c("Signficant Z-Score",   "Moran's I Z-Score")) +

并添加

scale_shape_discrete(name="", label=c("Signficant Z-Score", "Moran's I Z-Score")) +

(你也可以摆脱scale_fill_discrete,因为你实际上并没有在任何地方使用填充美学。)

这给了 enter image description here