14.2 Exploring

14.2.1 tabyl

tabyl() 的设计初衷是替代 Base R 中 table()，后者有几个缺点：

• 不接受数据框输入
  
• 不返回数据框
  
• 返回的结果很难进一步修饰

tabyl() 用于构建 1 ~ 3 个变量的（交叉）频数表，它 建立在 dplyr 和 tidyr 之上，所以以数据框基本输入、输出对象（但也可以接受一维向量），janitor 还提供了 adorn_* 函数族对其返回的表格进行修饰。以 starwars的一个子集演示 tabyl() 的用法：

humans <- starwars %>% 
  filter(species == "Human")

---

One-way tabyl

一维频数表

t1 <- humans %>%
  tabyl(eye_color)

t1
#>  eye_color  n percent
#>       blue 12  0.3429
#>  blue-gray  1  0.0286
#>      brown 17  0.4857
#>       dark  1  0.0286
#>      hazel  2  0.0571
#>     yellow  2  0.0571

tably() 可以聪明地处理数据中包含缺失值的情况：

x <- c("big", "big", "small", "small", "small", NA)

tabyl(x)
#>      x n percent valid_percent
#>    big 2   0.333           0.4
#>  small 3   0.500           0.6
#>   <NA> 1   0.167            NA
tabyl(x, show_na = F)
#>      x n percent
#>    big 2     0.4
#>  small 3     0.6

大部分 adorn_* 函数主要用于二维列联表，但也可以适用一维频数表：

t1 %>% 
  adorn_pct_formatting()
#>  eye_color  n percent
#>       blue 12   34.3%
#>  blue-gray  1    2.9%
#>      brown 17   48.6%
#>       dark  1    2.9%
#>      hazel  2    5.7%
#>     yellow  2    5.7%

---

Two-way tabyl

df %>% tabyl(var_1, var_2) 等同于 df %>% count(var_1, var_2) 后 pivot_wider() 展开其中的某一列，生成列联表：

t2 <- humans %>%
  tabyl(gender, eye_color)

t2
#>  gender blue blue-gray brown dark hazel yellow
#>  female    3         0     5    0     1      0
#>    male    9         1    12    1     1      2

# count() + pivot_wider()
humans %>% 
  count(gender, eye_color) %>% 
  pivot_wider(names_from = eye_color, values_from = n)
#> # A tibble: 2 x 7
#>   gender  blue brown hazel `blue-gray`  dark yellow
#>   <chr>  <int> <int> <int>       <int> <int>  <int>
#> 1 female     3     5     1          NA    NA     NA
#> 2 male       9    12     1           1     1      2

用于修饰的 adorn_* 函数有：

• adorn_totals(c("row", "col")): 添加行列汇总
  

• adorn_percentages(c("row", "col"))： 将交叉表的指替换为行或列百分比
  

• adorn_pct_formatting(digits, rounding): 决定百分比的格式
  

• adorn_rounding(): Round a data.frame of numbers (usually the result of adorn_percentages), either using the base R round() function or using janitor’s round_half_up() to round all ties up (thanks, StackOverflow).

  • e.g., round 10.5 up to 11, consistent with Excel’s tie-breaking behavior.
  • This contrasts with rounding 10.5 down to 10 as in base R’s round(10.5).
  • adorn_rounding() returns columns of class numeric, allowing for graphing, sorting, etc. It’s a less-aggressive substitute for adorn_pct_formatting(); these two functions should not be called together.

• adorn_ns(): add Ns to a tabyl. These can be drawn from the tabyl’s underlying counts, which are attached to the tabyl as metadata, or they can be supplied by the user.

• adorn_title(placement, row_name, col_name): “combined” 或者 “top”，调整行变量名称的位置

注意在应用这些帮助函数时要遵从一定的逻辑顺序。例如，adorn_ns() 和 adorn_percent_fomatting() 应该在调用 adorn_percentages() 之后。

对 t2 应用 adorn_* 函数：

t2 %>% 
  adorn_totals("col") %>% 
  adorn_percentages("row") %>%
  adorn_pct_formatting(digits = 2) %>% 
  adorn_ns() %>%
  adorn_title("combined")
#>  gender/eye_color       blue blue-gray       brown      dark      hazel
#>            female 33.33% (3) 0.00% (0) 55.56%  (5) 0.00% (0) 11.11% (1)
#>              male 34.62% (9) 3.85% (1) 46.15% (12) 3.85% (1)  3.85% (1)
#>     yellow        Total
#>  0.00% (0) 100.00%  (9)
#>  7.69% (2) 100.00% (26)

tabyl 对象最终可以传入 knitr::kabel() 中呈现

t2 %>% 
  adorn_totals("row") %>% 
  adorn_percentages("col") %>%
  adorn_pct_formatting(digits = 1) %>% 
  adorn_ns() %>%
  adorn_title("top", row_name = "gender", col_name = "color") %>%
  knitr::kable()

col	or
gender	blue	blue-gray	brown	dark	hazel	yellow
female	25.0% (3)	0.0% (0)	29.4% (5)	0.0% (0)	50.0% (1)	0.0% (0)
male	75.0% (9)	100.0% (1)	70.6% (12)	100.0% (1)	50.0% (1)	100.0% (2)
Total	100.0% (12)	100.0% (1)	100.0% (17)	100.0% (1)	100.0% (2)	100.0% (2)

---

Three-way tabyl

在 tabyl() 中传入三个变量时，返回一个二维 tabyl 的列表：

t3 <- humans %>%
  tabyl(eye_color, skin_color, gender)

t3
#> $female
#>  eye_color dark fair light pale tan white
#>       blue    0    2     1    0   0     0
#>  blue-gray    0    0     0    0   0     0
#>      brown    0    1     4    0   0     0
#>       dark    0    0     0    0   0     0
#>      hazel    0    0     1    0   0     0
#>     yellow    0    0     0    0   0     0
#> 
#> $male
#>  eye_color dark fair light pale tan white
#>       blue    0    7     2    0   0     0
#>  blue-gray    0    1     0    0   0     0
#>      brown    3    4     3    0   2     0
#>       dark    1    0     0    0   0     0
#>      hazel    0    1     0    0   0     0
#>     yellow    0    0     0    1   0     1

这时的 adorn_* 函数将会应用于列表中的每个 tabyl 元素：

t3 %>% 
  adorn_percentages("row") %>%
  adorn_pct_formatting(digits = 0) %>%
  adorn_ns()
#> $female
#>  eye_color   dark    fair    light   pale    tan  white
#>       blue 0% (0) 67% (2)  33% (1) 0% (0) 0% (0) 0% (0)
#>  blue-gray  - (0)   - (0)    - (0)  - (0)  - (0)  - (0)
#>      brown 0% (0) 20% (1)  80% (4) 0% (0) 0% (0) 0% (0)
#>       dark  - (0)   - (0)    - (0)  - (0)  - (0)  - (0)
#>      hazel 0% (0)  0% (0) 100% (1) 0% (0) 0% (0) 0% (0)
#>     yellow  - (0)   - (0)    - (0)  - (0)  - (0)  - (0)
#> 
#> $male
#>  eye_color     dark     fair   light    pale     tan   white
#>       blue   0% (0)  78% (7) 22% (2)  0% (0)  0% (0)  0% (0)
#>  blue-gray   0% (0) 100% (1)  0% (0)  0% (0)  0% (0)  0% (0)
#>      brown  25% (3)  33% (4) 25% (3)  0% (0) 17% (2)  0% (0)
#>       dark 100% (1)   0% (0)  0% (0)  0% (0)  0% (0)  0% (0)
#>      hazel   0% (0) 100% (1)  0% (0)  0% (0)  0% (0)  0% (0)
#>     yellow   0% (0)   0% (0)  0% (0) 50% (1)  0% (0) 50% (1)

14.2.2 get_dupes

get_dupes(dat, ...) 返回数据框dat中在变量...上重复的观测，以及重复的次数：

mtcars %>% 
  get_dupes(wt, cyl)
#> # A tibble: 4 x 12
#>      wt   cyl dupe_count   mpg  disp    hp  drat  qsec    vs    am  gear  carb
#>   <dbl> <dbl>      <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1  3.44     6          2  19.2  168.   123  3.92  18.3     1     0     4     4
#> 2  3.44     6          2  17.8  168.   123  3.92  18.9     1     0     4     4
#> 3  3.57     8          2  14.3  360    245  3.21  15.8     0     0     3     4
#> 4  3.57     8          2  15    301    335  3.54  14.6     0     1     5     8

14.2.3 remove_

14.2.3.1 remove_empty

remove_empty(c("rows", "cols")) 移除行或列（或行和列）上全为 NA 值的观测：

q <- data.frame(v1 = c(1, NA, 3),
                v2 = c(NA, NA, NA),
                v3 = c("a", NA, "b"))
q %>%
  remove_empty(c("rows", "cols"))
#>   v1 v3
#> 1  1  a
#> 3  3  b

q %>% 
  remove_empty("rows")
#>   v1 v2 v3
#> 1  1 NA  a
#> 3  3 NA  b

q %>% 
  remove_empty("cols")
#>   v1   v3
#> 1  1    a
#> 2 NA <NA>
#> 3  3    b

remove_empty 的实现原理很简单，以移除空的行观测为例：如果某行全为 NA，则该行对应的 rowSums(is.na(dat)) = ncol(dat):

function (dat, which = c("rows", "cols")) 
{
    if (missing(which) && !missing(dat)) {
        message("value for \"which\" not specified, defaulting to c(\"rows\", \"cols\")")
        which <- c("rows", "cols")
    }
    if ((sum(which %in% c("rows", "cols")) != length(which)) && 
        !missing(dat)) {
        stop("\"which\" must be one of \"rows\", \"cols\", or c(\"rows\", \"cols\")")
    }
    if ("rows" %in% which) {
        dat <- dat[rowSums(is.na(dat)) != ncol(dat), , drop = FALSE]
    }
    if ("cols" %in% which) {
        dat <- dat[, colSums(!is.na(dat)) > 0, drop = FALSE]
    }
    dat
}

14.2.3.2 remove_constant

remove_constant() 移除数据框中的常数列：

a <- data.frame(good = 1:3, boring = "the same")
a %>% remove_constant()
#>   good
#> 1    1
#> 2    2
#> 3    3

14.2.4 round_half_up

Base R 中的取整函数 round() 采取的规则是 “四舍六入五留双”（Banker’s Rounding，当小数位是 .5 时，若前一位是奇数，则进 1 ； 若前一位数偶数，则退一）：

nums <- c(2.5, 3.5)
round(nums)
#> [1] 2 4

round_half_up 遵循最简单的四舍五入规则:

round_half_up(nums)
#> [1] 3 4

若希望取整到特定的小数位，例如 0, 0.25, 0.5, 0.75, 1。可以用 round_half_fraction() 并指定除数

14.2.5 excel_numeric_to_date

excel_numeric_to_date() 按照 Excel 编码日期的规则(1989/12/31 = 1) 将整数转换为数字：

excel_numeric_to_date(41103)
#> [1] "2012-07-13"
excel_numeric_to_date(41103.01) # ignores decimal places, returns Date object
#> [1] "2012-07-13"

14.2.6 top_levels

在李克特量表数据的分析中，常需要知道某个态度变量中占比最高的几个水平，这样的变量在 R 中以有序因子的方式储存，top_levels() 将有序因子的所有水平分为三组（左，中间，右），并分别呈现各组的频数：

f <- factor(c("strongly agree", "agree", "neutral", "neutral", "disagree", "strongly agree"),
            levels = c("strongly agree", "agree", "neutral", "disagree", "strongly disagree"))
top_levels(f)
#>                            f n percent
#>        strongly agree, agree 3   0.500
#>                      neutral 2   0.333
#>  disagree, strongly disagree 1   0.167

top_levels(as.factor(mtcars$hp))
#>                  as.factor(mtcars$hp)  n percent
#>                                52, 62  2  0.0625
#>  <<< Middle Group (18 categories) >>> 28  0.8750
#>                              264, 335  2  0.0625

改变两侧分组包含水平的个数：

top_levels(as.factor(mtcars$hp), n = 4)
#>                  as.factor(mtcars$hp)  n percent
#>                        52, 62, 65, 66  5   0.156
#>  <<< Middle Group (14 categories) >>> 22   0.688
#>                    230, 245, 264, 335  5   0.156

14.2.7 row_to_names

row_to_names() 将某个观测行提升至列名：

dirt <- data.frame(X_1 = c(NA, "ID", 1:3),
           X_2 = c(NA, "Value", 4:6))

dirt
#>    X_1   X_2
#> 1 <NA>  <NA>
#> 2   ID Value
#> 3    1     4
#> 4    2     5
#> 5    3     6

dirt %>% 
  row_to_names(row_number = 2, remove_rows_above = F)  
#>     ID Value
#> 1 <NA>  <NA>
#> 3    1     4
#> 4    2     5
#> 5    3     6

dirt %>% 
  row_to_names(row_number = 2, remove_rows_above = T)  
#>   ID Value
#> 3  1     4
#> 4  2     5
#> 5  3     6