如何更好地从 ggplot2 创建具有多个变量的堆叠条形图? [英] How to better create stacked bar graphs with multiple variables from ggplot2?

问题描述

我经常需要制作堆叠条形图来比较变量，并且因为我在 R 中完成所有统计数据，所以我更喜欢使用 ggplot2 在 R 中完成我的所有图形.我想学习如何做两件事:

I often have to make stacked barplots to compare variables, and because I do all my stats in R, I prefer to do all my graphics in R with ggplot2. I would like to learn how to do two things:

首先，我希望能够为每个变量添加适当的百分比刻度线，而不是按计数添加刻度线.计数会令人困惑，这就是我完全去掉轴标签的原因.

First, I would like to be able to add proper percentage tick marks for each variable rather than tick marks by count. Counts would be confusing, which is why I take out the axis labels completely.

其次，必须有一种更简单的方法来重新组织我的数据以实现这一目标.这似乎是我应该能够在 ggplot2 中使用 plyR 本地做的事情，但是 plyR 的文档不是很清楚(我已经阅读了 ggplot2 书和在线 plyR 文档.

Second, there must be a simpler way to reorganize my data to make this happen. It seems like the sort of thing I should be able to do natively in ggplot2 with plyR, but the documentation for plyR is not very clear (and I have read both the ggplot2 book and the online plyR documentation.

我最好的图表如下所示，创建它的代码如下:

My best graph looks like this, the code to create it follows:

我用来获取它的 R 代码如下:

The R code I use to get it is the following:

library(epicalc)  

### recode the variables to factors ###
recode(c(int_newcoun, int_newneigh, int_neweur, int_newusa, int_neweco, int_newit, int_newen, int_newsp, int_newhr, int_newlit, int_newent, int_newrel, int_newhth, int_bapo, int_wopo, int_eupo, int_educ), c(1,2,3,4,5,6,7,8,9, NA), 
c('Very Interested','Somewhat Interested','Not Very Interested','Not At All interested',NA,NA,NA,NA,NA,NA))

### Combine recoded variables to a common vector
Interest1<-c(int_newcoun, int_newneigh, int_neweur, int_newusa, int_neweco, int_newit, int_newen, int_newsp, int_newhr, int_newlit, int_newent, int_newrel, int_newhth, int_bapo, int_wopo, int_eupo, int_educ)


### Create a second vector to label the first vector by original variable ###  
a1<-rep("News about Bangladesh", length(int_newcoun))
a2<-rep("Neighboring Countries", length(int_newneigh))
[...]
a17<-rep("Education", length(int_educ))


Interest2<-c(a1, a2, a3, a4, a5, a6, a7, a8, a9, a10, a11, a12, a13, a14, a15, a16, a17)

### Create a Weighting vector of the proper length ###
Interest.weight<-rep(weight, 17)

### Make and save a new data frame from the three vectors ###
Interest.df<-cbind(Interest1, Interest2, Interest.weight)
Interest.df<-as.data.frame(Interest.df)

write.csv(Interest.df, 'C:\Documents and Settings\[name]\Desktop\Sweave\InterestBangladesh.csv')

### Sort the factor levels to display properly ###

Interest.df$Interest1<-relevel(Interest$Interest1, ref='Not Very Interested')
Interest.df$Interest1<-relevel(Interest$Interest1, ref='Somewhat Interested')
Interest.df$Interest1<-relevel(Interest$Interest1, ref='Very Interested')

Interest.df$Interest2<-relevel(Interest$Interest2, ref='News about Bangladesh')
Interest.df$Interest2<-relevel(Interest$Interest2, ref='Education')
[...]
Interest.df$Interest2<-relevel(Interest$Interest2, ref='European Politics')

detach(Interest)
attach(Interest)

### Finally create the graph in ggplot2 ###

library(ggplot2)
p<-ggplot(Interest, aes(Interest2, ..count..))
p<-p+geom_bar((aes(weight=Interest.weight, fill=Interest1)))
p<-p+coord_flip()
p<-p+scale_y_continuous("", breaks=NA)
p<-p+scale_fill_manual(value = rev(brewer.pal(5, "Purples")))
p
update_labels(p, list(fill='', x='', y=''))

我非常感谢任何提示、技巧或提示.

I'd very much appreciate any tips, tricks or hints.

推荐答案

你不需要 prop.tables 或 count 等来做 100% 堆叠的条形图.你只需要 +geom_bar(position="stack")

You don't need prop.tables or count etc to do the 100% stacked bars. You just need +geom_bar(position="stack")

这篇关于如何更好地从 ggplot2 创建具有多个变量的堆叠条形图?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！