Chapter 4 R语言ggplot2柱形图

4.1 最基本的柱形图需要准备的数据

一列x一列y

• 如果柱子垂直 x是离散型数据 y是连续型数据

• 如果想要水平的柱子，就把y设置成离散数据，x设置成连续数据

• 数据集

var1	var2
A	1
B	2
C	3
D	4
E	5

读取数据集

library(readxl)
dat01<-read_excel("example_data/04-barplot/dat01.xlsx")
head(dat01)

## # A tibble: 5 x 2
##   var1   var2
##   <chr> <dbl>
## 1 A         1
## 2 B         2
## 3 C         3
## 4 D         4
## 5 E         5

作图代码

柱形图的函数有geom_col()和geom_bar(),具体有什么区别我没有仔细研究过，我自己习惯用geom_col()函数，做堆积柱形图和簇状柱形图的时候会使用geom_bar()函数

柱形图可以修改的参数分别是

• color对应柱子的边框颜色
• size 对应是边框的粗细
• fill对应柱子的填充颜色
• alpha对应的是柱子填充颜色的透明度，取值是0到1之间
• width对应柱子的宽度

看如下代码的效果，你可以试着更改每个参数的值

library(readxl)
dat01<-read_excel("example_data/04-barplot/dat01.xlsx")
head(dat01)

## # A tibble: 5 x 2
##   var1   var2
##   <chr> <dbl>
## 1 A         1
## 2 B         2
## 3 C         3
## 4 D         4
## 5 E         5

library(ggplot2)
ggplot(data=dat01,aes(x=var1,y=var2))+
  geom_col(color="red",size=3,
           fill="blue",alpha=0.5,
           width = 0.2)

如果统一修改边框颜色，填充颜色这些属性，比如上面的例子，5个柱子全都设置成一样的，就把参数设置到aes()的外面。如果想要用数据中的某一列来映射颜色。需要把参数写到aes()的里面。

柱形图还有一个比较常用的操作是添加误差线，这里假设已经算好了标准差，我们将标准差整理到数据集里，格式如下

添加误差线的函数是geom_errorbar()

如果是垂直与x轴的误差线，需要制定ymin和ymax两个参数

作图代码

library(readxl)
dat01<-read_excel("example_data/04-barplot/dat01_1.xlsx")
head(dat01)

## # A tibble: 5 x 3
##   var1   var2 sd_value
##   <chr> <dbl>    <dbl>
## 1 A         1      0.5
## 2 B         2      0.5
## 3 C         3      0.5
## 4 D         4      0.5
## 5 E         5      0.5

library(ggplot2)
ggplot(data=dat01,aes(x=var1,y=var2))+
  geom_col()+
  geom_errorbar(aes(ymin=var2-sd_value,
                    ymax=var2+sd_value))

误差线函数比较常用的三个参数是

• width 调节误差线的宽度
• color 误差线的颜色
• lty 线的类型，就是实线 虚线 这些

更改这三个参数

library(readxl)
dat01<-read_excel("example_data/04-barplot/dat01_1.xlsx")
head(dat01)

## # A tibble: 5 x 3
##   var1   var2 sd_value
##   <chr> <dbl>    <dbl>
## 1 A         1      0.5
## 2 B         2      0.5
## 3 C         3      0.5
## 4 D         4      0.5
## 5 E         5      0.5

library(ggplot2)
ggplot(data=dat01,aes(x=var1,y=var2))+
  geom_col(fill="grey")+
  geom_errorbar(aes(ymin=var2-sd_value,
                    ymax=var2+sd_value),
                color="red",
                width=0.4,
                lty="dashed")

柱形图还有一个比较常用的操作是组间比较，做统计检验后添加p值或者显著性的星号，这里还是假设已经做好了统计检验，这里使用annotate()函数来添加线段和文本。这里有一个知识点是如果是离散数据作为x轴，第一个柱子的横坐标是1,第二个第三个依次是2，和3 这样

首先是annotate()函数添加添加文本，需要制定四个内容，

• 添加注释的类型 文本是geom = “text”
• 添加文本的位置 一个 x 和一个 y
• 添加文本的内容 label = “ABC”

如下代码 我在 A B 两个柱子中间 横坐标是1.5 纵坐标3的位置添加一个P value = 0.01的文本,设置文本的颜色为蓝色

library(readxl)
dat01<-read_excel("example_data/04-barplot/dat01_1.xlsx")
head(dat01)

## # A tibble: 5 x 3
##   var1   var2 sd_value
##   <chr> <dbl>    <dbl>
## 1 A         1      0.5
## 2 B         2      0.5
## 3 C         3      0.5
## 4 D         4      0.5
## 5 E         5      0.5

library(ggplot2)
ggplot(data=dat01,aes(x=var1,y=var2))+
  geom_col(fill="grey")+
  geom_errorbar(aes(ymin=var2-sd_value,
                    ymax=var2+sd_value),
                color="red",
                width=0.4,
                lty="dashed")+
  annotate(geom = "text",x=1.5,y=3,
           label="P value = 0.01",color="blue")

接下来是添加注释的线段，线段需要制定的参数是

• geom=“segment”

• 线段的起始位置 x,y 线段的终止位置x y

• 还可以更改颜色 线型 粗细 之类的

看如下代码

library(readxl)
dat01<-read_excel("example_data/04-barplot/dat01_1.xlsx")
head(dat01)

## # A tibble: 5 x 3
##   var1   var2 sd_value
##   <chr> <dbl>    <dbl>
## 1 A         1      0.5
## 2 B         2      0.5
## 3 C         3      0.5
## 4 D         4      0.5
## 5 E         5      0.5

library(ggplot2)
ggplot(data=dat01,aes(x=var1,y=var2))+
  geom_col(fill="grey")+
  geom_errorbar(aes(ymin=var2-sd_value,
                    ymax=var2+sd_value),
                color="red",
                width=0.4,
                lty="dashed")+
  annotate(geom = "text",x=1.5,y=3,
           label="P value = 0.01",color="blue")+
  annotate(geom = "segment",x=1,y=1.5,xend=1,yend=2.8,
           color="black",size=1)

接下来再来添加两个线段

library(readxl)
dat01<-read_excel("example_data/04-barplot/dat01_1.xlsx")
head(dat01)

## # A tibble: 5 x 3
##   var1   var2 sd_value
##   <chr> <dbl>    <dbl>
## 1 A         1      0.5
## 2 B         2      0.5
## 3 C         3      0.5
## 4 D         4      0.5
## 5 E         5      0.5

library(ggplot2)
ggplot(data=dat01,aes(x=var1,y=var2))+
  geom_col(fill="grey")+
  geom_errorbar(aes(ymin=var2-sd_value,
                    ymax=var2+sd_value),
                color="red",
                width=0.4,
                lty="dashed")+
  annotate(geom = "text",x=1.5,y=3,
           label="P value = 0.01",color="blue")+
  annotate(geom = "segment",x=1,y=1.5,xend=1,yend=2.8,
           color="black",size=1)+
  annotate(geom = "segment",x=2,y=2.5,xend=2,yend=2.8,
           color="black",size=1)+
    annotate(geom = "segment",x=1,y=2.8,xend=2,yend=2.8,
           color="black",size=1)

用这个annotate()函数相对来说会比较繁琐，如果要添加比较多的线段的话代码会很长，但是这个可定制性比较强，我目前还想不到其他比较合适的替代方案

4.2 簇状柱形图

参考链接 https://r-graph-gallery.com/48-grouped-barplot-with-ggplot2.html

数据格式如下

var1	var2	value
group1	A	13.411067
group1	B	6.884985
group1	C	37.514516
group2	A	25.970176
group2	B	12.292093
group2	C	10.388148
group3	A	3.411084
group3	B	2.060476
group3	C	9.582556
group4	A	4.190190
group4	B	11.724548
group4	C	3.871083

作图代码

library(readxl)
dat01<-read_excel("example_data/04-barplot/dat02_grouped_barplot.xlsx")
library(ggplot2)
ggplot(data=dat01,aes(x=var1,y=value))+
  geom_bar(stat="identity",
           aes(fill=var2),
           position = "dodge")

• 这里stat="identity"我一直没有搞清楚是什么意思,记住是必须要写的

• 如果不加position="dodge"默认是堆积柱形图

• 堆积柱形图的position应该设置为position="stack"

簇状柱形图比较常用的修改参数是

• 不同组之前的显示顺序，默认是首字母
• 组内不同柱子的的排序，默认也是首字母

涉及到顺序的都是调节数据集的因子水平

代码

library(readxl)
dat01<-read_excel("example_data/04-barplot/dat02_grouped_barplot.xlsx")
dat01$var1<-factor(dat01$var1,
                   levels = c("group3","group2","group4","group1"))
library(ggplot2)
ggplot(data=dat01,aes(x=var1,y=value))+
  geom_bar(stat="identity",
           aes(fill=var2),
           position = "dodge")

dat01$var2<-factor(dat01$var2,
                   levels = c("B","C","A"))
ggplot(data=dat01,aes(x=var1,y=value))+
  geom_bar(stat="identity",
           aes(fill=var2),
           position = "dodge")

堆积柱形图还有一个经常会遇到的问题是添加误差线，现在假设我们已经把标准差算好了，整理excel里，数据格式如下，在原有数据的基础上添加一列标准差的数据

var1	var2	value	sd_value
group1	A	13.411067	2
group1	B	6.884985	1
group1	C	37.514516	3
group2	A	25.970176	2
group2	B	12.292093	1
group2	C	10.388148	3
group3	A	3.411084	2
group3	B	2.060476	1
group3	C	9.582556	3
group4	A	4.190190	2
group4	B	11.724548	1
group4	C	3.871083	3

• 添加误差线的函数是 geom_errorbar()

作图代码

library(readxl)
dat01<-read_excel("example_data/04-barplot/dat02_grouped_barplot_01.xlsx")
library(ggplot2)
ggplot(data=dat01,aes(x=var1,y=value))+
  geom_bar(stat="identity",
           aes(fill=var2),
           position = "dodge")+
  geom_errorbar(aes(ymin=value-sd_value,
                    ymax=value+sd_value))

• 误差线主要的调节参数就两个，一个是width误差线的宽度，一个是color颜色

library(readxl)
dat01<-read_excel("example_data/04-barplot/dat02_grouped_barplot_01.xlsx")
library(ggplot2)
ggplot(data=dat01,aes(x=var1,y=value))+
  geom_bar(stat="identity",
           aes(fill=var2),
           position = "dodge")+
  geom_errorbar(aes(ymin=value-sd_value,
                    ymax=value+sd_value),
                width=0.4,
                color="grey")

簇状柱形图的误差线全部集中在同一位置，需要我们用参数position = position_dodge(1)调节开,这里需要注意一点是如果调节误差线的位置，需要把fill=var2参数写到ggplot里，position_dodge()里面的数值具体应该设置多少我也搞不清楚，每次都要设置好几次

library(readxl)
dat01<-read_excel("example_data/04-barplot/dat02_grouped_barplot_01.xlsx")
library(ggplot2)
ggplot(data=dat01,aes(x=var1,y=value,fill=var2))+
  geom_bar(stat="identity",
           position = "dodge")+
  geom_errorbar(aes(ymin=value-sd_value,
                    ymax=value+sd_value),
                width=0.4,
                color="blue",
                position = position_dodge(0.9))

其他的美化，比如让柱子贴着底，坐标轴标签，更改默认配色等

library(readxl)
dat01<-read_excel("example_data/04-barplot/dat02_grouped_barplot_01.xlsx")
library(ggplot2)
ggplot(data=dat01,aes(x=var1,y=value,fill=var2))+
  geom_bar(stat="identity",
           position = "dodge")+
  geom_errorbar(aes(ymin=value-sd_value,
                    ymax=value+sd_value),
                width=0.4,
                color="blue",
                position = position_dodge(0.9))+
  scale_y_continuous(expand=expansion(mult=c(0,0.1)))+
  scale_fill_manual(values = c("A"="grey","B"="black","C"="green"))+
  labs(x="AAAAA",y="BBBBB")

4.2.1 接下来看每周一图里面的例子

这里 aes()的内容是可以写到作图函数里，也可以写到ggplot里，这里还是有区别的，比如上面提到的误差线的位置调节

library(readr)
success_rates<-read_csv("example_data/04-barplot/success_rates.csv")
library(ggplot2)
ggplot(success_rates) +
  # add bar for each discipline colored by gender
  geom_bar(aes(x = discipline, y = success, fill = gender),
           stat = "identity", position = "dodge") +
  # name axes and remove gap between bars and y-axis
  scale_y_continuous("Success Rate", expand = c(0, 0)) +
  scale_x_discrete("Discipline") +
  scale_fill_manual(values = c("#468189", "#9DBEBB")) +
  # remove grey theme
  theme_classic(base_size = 18) +
  # rotate x-axis and remove superfluous axis elements
  theme(axis.text.x = element_text(angle = 90, 
                                   hjust = 1, vjust = 0),
        axis.line = element_blank(),
        axis.ticks.x = element_blank()) 

4.3 堆积柱形图

堆积柱形图和簇状柱形图的数据格式是一样的，自己的数据具体需要用堆积柱形图还是簇状柱形图自己斟酌，堆积柱形图我们只需要把簇状柱形图对应的position="dodge"改成position="stack"就可以了

dat01<-read_excel("example_data/04-barplot/dat02_grouped_barplot.xlsx")
library(ggplot2)
ggplot(data=dat01,aes(x=var1,y=value))+
  geom_bar(stat="identity",
           aes(fill=var2),
           position = "stack")

堆积柱形图有一个特点是除了展示真实数据外，还可以展示比例,需要我们把position="stack"改成position="fill"

dat01<-read_excel("example_data/04-barplot/dat02_grouped_barplot.xlsx")
library(ggplot2)
ggplot(data=dat01,aes(x=var1,y=value))+
  geom_bar(stat="identity",
           aes(fill=var2),
           position = "fill")

接下来就是更改柱子的顺序，和簇状柱形图调节顺序一样，只要更改原始数据的因子水平,默认的顺序是从上往下排的

library(readxl)
dat01<-read_excel("example_data/04-barplot/dat02_grouped_barplot.xlsx")
dat01$var1<-factor(dat01$var1,
                   levels = c("group3","group2","group4","group1"))
library(ggplot2)
ggplot(data=dat01,aes(x=var1,y=value))+
  geom_bar(stat="identity",
           aes(fill=var2),
           position = "stack")

dat01$var2<-factor(dat01$var2,
                   levels = c("B","C","A"))
ggplot(data=dat01,aes(x=var1,y=value))+
  geom_bar(stat="identity",
           aes(fill=var2),
           position = "stack")

• 堆积柱形图添加误差线不常用，但也有人有这个需求，需要对原始数据有一个累加处理用来指定误差线的y坐标

library(readxl)
dat01<-read_excel("example_data/04-barplot/dat02_grouped_barplot_01.xlsx")
library(tidyverse)
dat01 %>% 
  group_by(var1) %>% 
  mutate(new_col=cumsum(value)) -> dat01

readr::write_csv(dat01,file="example_data/04-barplot/dat02_grouped_barplot_01.csv")
dat01$var2<-factor(dat01$var2,
                   levels = c("C","B","A"))
ggplot(data=dat01,aes(x=var1,y=value))+
  geom_bar(stat="identity",
           aes(fill=var2),
           position = "stack")+
  geom_errorbar(aes(ymin=new_col-sd_value,
                    ymax=new_col+sd_value),
                width=0.4,
                color="grey")

堆积柱形图还有一个经常遇到的问题是在图上添加文字，我们自数据集里添加新的列指定文本标签和文本标签的坐标,

library(readxl)
dat01<-read_excel("example_data/04-barplot/dat02_grouped_barplot_02.xlsx")
knitr::kable(dat01, "simple")

var1	var2	value	sd_value	text_y	text
group1	A	13.411067	2	45	a
group1	B	6.884985	1	40	b
group1	C	37.514516	3	20	c
group2	A	25.970176	2	40	d
group2	B	12.292093	1	18	e
group2	C	10.388148	3	5	f
group3	A	3.411084	2	12	g
group3	B	2.060476	1	11	h
group3	C	9.582556	3	5	i
group4	A	4.190190	2	18	j
group4	B	11.724548	1	10	k
group4	C	3.871083	3	2	l

添加文本用到的函数是geom_text()

dat01<-read_excel("example_data/04-barplot/dat02_grouped_barplot_02.xlsx")
library(ggplot2)
ggplot(data=dat01,aes(x=var1,y=value))+
  geom_bar(stat="identity",
           aes(fill=var2),
           position = "stack")+
  geom_text(aes(x=var1,y=text_y,label=text))

美化 更改配色

4.3.1 实际例子

library(readr)
library(tidyverse)

bechdel_test_df<-read_csv("example_data/04-barplot/bechdel_test_df.csv")

## Rows: 43 Columns: 4
## -- Column specification ---------------------------------------------------------------------
## Delimiter: ","
## chr (2): year_group, clean_test
## dbl (2): category_count, category_prop
## 
## i Use `spec()` to retrieve the full column specification for this data.
## i Specify the column types or set `show_col_types = FALSE` to quiet this message.

bechdel_test_text <- read_csv("example_data/04-barplot/bechdel_test_text.csv")

## Rows: 5 Columns: 8
## -- Column specification ---------------------------------------------------------------------
## Delimiter: ","
## chr (3): year_group, clean_test, label
## dbl (5): category_count, category_prop, prop_cum, prop_cum_lag, y
## 
## i Use `spec()` to retrieve the full column specification for this data.
## i Specify the column types or set `show_col_types = FALSE` to quiet this message.

bechdel_step_df <- read_csv("example_data/04-barplot/bechdel_step_df.csv")

## Rows: 9 Columns: 6
## -- Column specification ---------------------------------------------------------------------
## Delimiter: ","
## chr (1): year_group
## dbl (5): category_prop, x_coord, x_end_coord, y_coord, y_end_coord
## 
## i Use `spec()` to retrieve the full column specification for this data.
## i Specify the column types or set `show_col_types = FALSE` to quiet this message.

library(ggplot2)

bechdel_test_df %>% 
  mutate(clean_test = factor(clean_test, 
                             levels = c("ok", "dubious", "men", "notalk", "nowomen")),
         clean_test = fct_rev(clean_test)) %>% 
  ggplot(aes(year_group, category_prop)) +
  geom_col(aes(fill = clean_test), width = 1, color = "white", 
           size = 0.6, show.legend = FALSE) +
  geom_segment(data = bechdel_step_df, 
               aes(x = x_coord, xend = x_end_coord, 
                   y = category_prop, yend = category_prop), 
               size = 1.5) +
  geom_segment(data = filter(bechdel_step_df, year_group != "2010 -\n'13"), 
               aes(x = x_end_coord, xend = x_end_coord, 
                   y = y_coord, yend = y_end_coord), 
               lineend = "round", size = 1.5) +
  geom_segment(aes(x = 0.12, xend = 9.5, y = 0, yend = 0), 
               size = 0.8) +
  geom_segment(aes(x = 0.25, xend = 9.5, y = 1, yend = 1),
               size = 0.8, color = "#cdcdcd") +
  geom_segment(data = tibble(x = 0.12, xend = 0.5, y = c(0.25, 0.5, 0.75)), 
               aes(x = x, xend = xend, y = y, yend = y), 
               size = 0.8, color = "#cdcdcd") +
  geom_text(data = tibble(x = 0, y = c(0, 0.25, 0.5, 0.75, 1), 
                          label = c(0, 25, 50, 75, 100)), 
            aes(x = x, y = y, label = label), 
            family = "serif", size = 6, hjust = 1) +
  geom_text(data = tibble(x = 0.2, y = 1, label = "%"), 
            aes(x = x, y = y, label = label), 
            family = "serif", size = 7, hjust = 1) +
  geom_text(data = tibble(x = c(0.5, 2.5, 4.5, 6.5, 8.5), 
                          y = -0.06, 
                          label = c("1970-\n'74", "1980-\n'84", "1990-\n'94", "2000-\n'04", "2010-\n'13")), 
            aes(x = x, y = y, label = label), 
            family = "serif", size = 6, hjust = -0.2, lineheight = 0.55) +
  geom_segment(data = bechdel_test_text, 
               aes(x = 9.5, xend = 9.75, y = y, yend = y), 
               size = 0.8) +
  geom_text(data = bechdel_test_text, 
            aes(x = 9.85, y = y, label = label), 
            family = "serif", hjust = 0, 
            vjust = 0.5, size = 6, lineheight = 0.6) +
  annotate("text", x = 3.75, y = 0.22, 
           label = "PASS", family = "serif", 
           fontface = "bold", size = 25, 
           hjust = 0, vjust = 0.5) +
  annotate("text", x = 4, y = 0.75, 
           label = "FAIL", family = "serif",
           fontface = "bold", size = 25, hjust = 0, vjust = 0.5) +
  scale_y_continuous(expand = c(0, 0)) +
  scale_fill_manual(values = c("ok" = "#008fd5", "dubious" = "#6bb2d5",
                               "men" = "#ffc9bf", "notalk" = "#ff9380", "nowomen" = "#ff2700")) +
  labs(title = "The Bechdel Test Over Time", 
       subtitle = "How women are represented in movies", 
       x = "", y = "", 
       caption = "Original plot by Fivethirtyeight | Replicated in R by Kaustav Sen") +
  coord_cartesian(clip = "off") +
  theme_void() +
  theme(
    plot.title.position = "plot",
    plot.title = element_text(family = "serif", face = "bold", 
                              size = 30, hjust = -0.12, margin = margin(b = 5)),
    plot.subtitle = element_text(family = "serif", size = 24, hjust = -0.12, margin = margin(b = 25)),
    plot.caption = element_text(family = "serif", size = 14, hjust = 0.5, vjust = -25, color = "grey70"),
    plot.margin = margin(20, 90, 25, 45),
    plot.background = element_rect(fill = "#f0f0f0", color = "#f0f0f0")
  )