ggplot2显示错误的条形图(每12个条形图为最小值)

Question

我正在尝试用下面的数据集用ggplot2制作条形图。

Incentive Marginal.Abatement.Cost    Tech
1  Status Quo                    -100  Tech 1
2  Status Quo                     -99  Tech 2
3  Status Quo                     -98  Tech 3
4  Status Quo                     -97  Tech 4
5  Status Quo                     -96  Tech 5
6  Status Quo                     -95  Tech 6
7  Status Quo                     -94  Tech 7
8  Status Quo                     -93  Tech 8
9  Status Quo                     -92  Tech 9
10 Status Quo                     -91 Tech 10
11 Status Quo                     -90 Tech 11
12 Status Quo                     -89 Tech 12
13 Status Quo                     -88 Tech 13
14 Status Quo                     -87 Tech 14
15 Status Quo                     -85 Tech 15
16 Status Quo                     -83 Tech 16
17 Status Quo                     -81 Tech 17
18 Status Quo                     -79 Tech 18
19 Status Quo                     -77 Tech 19
20 Status Quo                     -75 Tech 20
21 Status Quo                     -73 Tech 21

对于91个观察值，数据集就是这样的。当我使用下面的ggplot代码绘制它时，由于某些原因，每12个条形图都会显示-100，而不管它应该是什么。

ggplot(CRIFQuo, aes(x = Tech, y = Marginal.Abatement.Cost)) + geom_bar(stat = "identity")

​

​

Answer 1

正如评论者所确定的那样，R默认按字母顺序对因子级别进行排序，例如

sort(paste0("x",1:12))
## [1] "x1"  "x10" "x11" "x12" "x2"  "x3"  "x4"  "x5"  ...

另一方面，

gtools::mixedsort(paste0("x",1:12))
## [1] "x1"  "x2"  "x3"  "x4"  "x5"  "x6"  "x7"  ...

您可以重新排序因子级别，如下所示

CRIFQuo$x <- with(CRIFQuo,factor(x,levels=gtools::mixedsort(levels(x)))

在重新尝试绘图之前(当然，您需要首先安装gtools包)

或者，您可以将x变量转换为数字：

CRIFQuo$nTech <- as.numeric(gsub("^Tech ","",CRIFQuo$Tech)
ggplot(CRIFQuo, aes(x = nTech, y = Marginal.Abatement.Cost)) ...