r - Dotchart(在 base R 中)或 ggplot2 中具有分组数据的等价物

Question

我找不到在点图中按组绘制数据的方法(或 ggplot2 中的任何等效绘图类型)。示例数据如下所示，分组变量为 Turn_no:

df
# A tibble: 8 × 4
# Groups:   Turn_no [4]
  c5    Turn_no Word       RelPosition
  <chr>   <int> <chr>            <dbl>
1 NN1         9 daddy            0.111
2 NN2         9 apartments       0.778
3 NN1        13 job              1    
4 NN1        15 bit              0.167
5 NP0        15 mark             0.667
6 NP0        16 michael          0.222
7 NN1        16 gel              0.778
8 NN1        16 morning          1  

我想做的是绘制按 Turn_no 分组的 RelPosition 值，以便每行有与 RelPosition 值一样多的点每个 Turn_no 组。

我尝试将 dplyr 语法与 (base R) dotchart 结合起来的尝试失败了:

libtary(tidyverse)
df %>%
  group_by(Turn_no) %>%
  dotchart(RelPosition)
Error in dotchart(., RelPosition) : 
  'x' must be a numeric vector or matrix

我所能做的就是这样，但是不管它属于哪个组，它都会将每个 RelPosition 值放在自己的行上:

dotchart(df$RelPosition)

grouped dot plot 如何在 base R 中或在 ggplot2 中实现(geom_dotplot 似乎不等同于 点图)?

数据:

df <- structure(list(c5 = c("NN1", "NN2", "NN1", "NN1", "NP0", "NP0", 
                            "NN1", "NN1"), Turn_no = c(9L, 9L, 13L, 15L, 15L, 16L, 16L, 16L
                            ), Word = c("daddy", "apartments", "job", "bit", "mark", "michael", 
                                        "gel", "morning"), RelPosition = c(0.111111111111111, 0.777777777777778, 
                                                                           1, 0.166666666666667, 0.666666666666667, 0.222222222222222, 0.777777777777778, 
                                                                           1)), class = c("grouped_df", "tbl_df", "tbl", "data.frame"), row.names = c(NA, 
                                                                                                                                                      -8L), groups = structure(list(Turn_no = c(9L, 13L, 15L, 16L), 
                                                                                                                                                                                    .rows = structure(list(1:2, 3L, 4:5, 6:8), ptype = integer(0), class = c("vctrs_list_of", 
                                                                                                                                                                                                                                                             "vctrs_vctr", "list"))), row.names = c(NA, -4L), class = c("tbl_df", 
                                                                                                                                                                                                                                                                                                                        "tbl", "data.frame"), .drop = TRUE))

Answer 1

是这样的吗？

library(ggplot2)

df <- structure(list(c5 = c(
  "NN1", "NN2", "NN1", "NN1", "NP0", "NP0",
  "NN1", "NN1"
), Turn_no = c(9L, 9L, 13L, 15L, 15L, 16L, 16L, 16L), Word = c(
  "daddy", "apartments", "job", "bit", "mark", "michael",
  "gel", "morning"
), RelPosition = c(
  0.111111111111111, 0.777777777777778,
  1, 0.166666666666667, 0.666666666666667, 0.222222222222222, 0.777777777777778,
  1
)), class = c("grouped_df", "tbl_df", "tbl", "data.frame"), row.names = c(
  NA,
  -8L
), groups = structure(list(
  Turn_no = c(9L, 13L, 15L, 16L),
  .rows = structure(list(1:2, 3L, 4:5, 6:8), ptype = integer(0), class = c(
    "vctrs_list_of",
    "vctrs_vctr", "list"
  ))
), row.names = c(NA, -4L), class = c(
  "tbl_df",
  "tbl", "data.frame"
), .drop = TRUE))



df |>
  ggplot(aes(y = Turn_no, x = RelPosition, group = Turn_no)) +
  geom_line() +
  geom_point(size = 4)

^{由 reprex package 创建于 2023-01-07 (v2.0.1)}