Question

我正在尝试创建一个总结我的数据的图片。数据是关于从不同国家的不同实践中获得的药物使用的普遍程度。每个练习都贡献了不同数量的数据，我想在我的图片中显示所有这些。

以下是要处理的数据的子集：

gr<-data.frame(matrix(0,36))
gr$drug<-c("a","a","a","a","a","a","a","a","a","a","a","a","a","a","a","a","a","a","b","b","b","b","b","b","b","b","b","b","b","b","b","b","b","b","b","b")
gr$practice<-c("a","b","c","d","e","f","g","h","i","j","k","l","m","n","o","p","q","r","a","b","c","d","e","f","g","h","i","j","k","l","m","n","o","p","q","r")
gr$country<-c("c1","c1","c1","c1","c1","c1","c1","c1","c1","c1","c2","c2","c2","c2","c2","c2","c3","c3","c1","c1","c1","c1","c1","c1","c1","c1","c1","c1","c2","c2","c2","c2","c2","c2","c3","c3")
gr$prevalence<-c(9.14,5.53,16.74,1.93,8.51,14.96,18.90,11.18,15.00,20.10,24.56,22.29,19.41,20.25,25.01,25.87,29.33,20.76,18.94,24.60,26.51,13.37,23.84,21.82,23.69,20.56,30.53,16.66,28.71,23.83,21.16,24.66,26.42,27.38,32.46,25.34)
gr$prop<-c(0.027,0.023,0.002,0.500,0.011,0.185,0.097,0.067,0.066,0.023,0.433,0.117,0.053,0.199,0.098,0.100,0.594,0.406,0.027,0.023,0.002,0.500,0.011,0.185,0.097,0.067,0.066,0.023,0.433,0.117,0.053,0.199,0.098,0.100,0.594,0.406)
gr$low.CI<-c(8.27,4.80,12.35,1.83,7.22,14.53,18.25,10.56,14.28,18.76,24.25,21.72,18.62,19.83,24.36,25.22,28.80,20.20,17.73,23.15,21.06,13.12,21.79,21.32,22.99,19.76,29.60,15.41,28.39,23.25,20.34,24.20,25.76,26.72,31.92,24.73)
gr$high.CI<-c(10.10,6.37,22.31,2.04,10.00,15.40,19.56,11.83,15.74,21.52,24.87,22.86,20.23,20.68,25.67,26.53,29.86,21.34,20.21,26.10,32.79,13.63,26.02,22.33,24.41,21.39,31.48,17.98,29.04,24.43,22.01,25.12,27.09,28.05,33.01,25.95)

我写的代码就是这个

p<-ggplot(data=gr, aes(x=factor(drug), y=as.numeric(gr$prevalence), ymax=max(high.CI),position="dodge",fill=practice,width=prop))
colour<-c(rep("gray79",10),rep("gray60",6),rep("gray39",2))
p + theme_bw()+
  geom_bar(stat="identity",position = position_dodge(0.9)) +
  labs(x="Drug",y="Prevalence") + 
  geom_errorbar(ymax=gr$high.CI,ymin=gr$low.CI,position=position_dodge(0.9),width=0.25,size=0.25,colour="black",aes(x=factor(drug), y=as.numeric(gr$prevalence), fill=practice)) +
  ggtitle("Drug usage by country and practice") +
  scale_fill_manual(values = colour)+ guides(fill=F)

我获得的数字就是这样一个，当我想要它们“躲闪”时，它们都在彼此之上。

我还获得以下警告：

ymax未定义：使用y调整位置警告信息： position_dodge需要不重叠的x区间

理想情况下，我会让每个酒吧彼此靠近，它们的误差条位于酒吧中间，全部由国家组织。

我是否应该关注警告（我显然不完全理解）？

我希望这是有道理的。我希望我足够接近，但我似乎没有去任何地方，一些帮助将不胜感激。

谢谢

Answer 1

ggplot的zip接受宽度参数，但默认情况下不会将它们整齐地排列在躲避位置。以下解决方法引用了解决方案here：

geom_bar()

Answer 2

您在此处尝试传达的信息很多 - 使用条形图比较药物A和药物B并考虑比例，您可以使用facet_grid函数。试试这个：

      colour<-c(rep("gray79",10),rep("gray60",6),rep("gray39",2))




      gr$drug <- paste("Drug", gr$drug)
      p<-ggplot(data=gr, aes(x=factor(practice), y=as.numeric(prevalence), 
                             ymax=high.CI,ymin = low.CI, 
                             position="dodge",fill=practice, width=prop))


        p + theme_bw()+ facet_grid(drug~country, scales="free")  +
        geom_bar(stat="identity") +
        labs(x="Practice",y="Prevalence") + 
        geom_errorbar(position=position_dodge(0.9), width=0.25,size=0.25,colour="black") +
        ggtitle("Drug usage by country and practice") +
        scale_fill_manual(values = colour)+ guides(fill=F)

C1国家的宽度太小，正如您所指出的那样，一家诊所非常有影响力。

此外，您可以使用ggplot（aes（...））指定您的美学，而不必重置它，并且不需要在ggplot调用中的aes函数中包含数据框对象名称。

在ggplot2中躲避列

2 个答案: