发出ifelse()命令，其中两个不同数据帧中的两列看起来相同，但不标识为相同

Question

我在ifelse()命令中遇到了问题，其中两个不同数据帧中的两个列看起来是相同的，但它们并不是相同的。我可以使用任何指导来解决这个问题，这样代码就可以相互比较数据帧，并产生适当的输出，而不必自己手工输入/输入文本。

以下是我的两个起始数据集，df_1和df_2

> df_1
                   DV_name
1  submission_time_minutes
2  submission_time_minutes
3                       WC
4                       WC
5         Analytic_z_score
6         Analytic_z_score
7            Clout_z_score
8            Clout_z_score
9        Authentic_z_score
10       Authentic_z_score
11            Tone_z_score
12            Tone_z_score
13 submission_time_minutes
14 submission_time_minutes
15                      WC
16                      WC
17        Analytic_z_score
18        Analytic_z_score
19           Clout_z_score
20           Clout_z_score
21       Authentic_z_score
22       Authentic_z_score
23            Tone_z_score
24            Tone_z_score
25 submission_time_minutes
26 submission_time_minutes
27                      WC
28                      WC
29        Analytic_z_score
30        Analytic_z_score
31           Clout_z_score
32           Clout_z_score
33       Authentic_z_score
34       Authentic_z_score
35            Tone_z_score
36            Tone_z_score
37 submission_time_minutes
38 submission_time_minutes
39                      WC
40                      WC
41        Analytic_z_score
42        Analytic_z_score
43           Clout_z_score
44           Clout_z_score
45       Authentic_z_score
46       Authentic_z_score
47            Tone_z_score
48            Tone_z_score
> df_2
        Variable_analyses             Variable_label
1 submission_time_minutes Submission time in minutes
2                      WC                 Word count
3        Analytic_z_score             Analytic score
4           Clout_z_score                Clout score
5       Authentic_z_score            Authentic score
6            Tone_z_score                 Tone score

我想要创建来自df_2$Variable_analyses的列df_2$Variable_analyses，它基于df_1$DV_name和df_2$Variable_analyses之间的匹配材料。

下面是实现这一目标的，它是成功的：

> ## long way
> 
> ### creates Variable_label
> # ---- NOTE: does not directly extract Variable_label from df_2 and insert it into df_1
> # ---- NOTE: based on df_1$Variable_label
> df_1$Variable_label <- 
+   ifelse(df_1$DV_name == "submission_time_minutes", "Submission time in minutes",
+          ifelse(df_1$DV_name == "WC", "Word count",
+                 ifelse(df_1$DV_name == "Analytic_z_score", "Analytic score",
+                        ifelse(df_1$DV_name == "Clout_z_score", "Clout score",
+                               ifelse(df_1$DV_name == "Authentic_z_score", "Authentic score",
+                                      ifelse(df_1$DV_name == "Tone_z_score", "Tone score", NA
+                                      ))))))
> 
> ### displays df
> # ---- NOTE: displays df with created variable in desired output form
> df_1
                   DV_name             Variable_label
1  submission_time_minutes Submission time in minutes
2  submission_time_minutes Submission time in minutes
3                       WC                 Word count
4                       WC                 Word count
5         Analytic_z_score             Analytic score
6         Analytic_z_score             Analytic score
7            Clout_z_score                Clout score
8            Clout_z_score                Clout score
9        Authentic_z_score            Authentic score
10       Authentic_z_score            Authentic score
11            Tone_z_score                 Tone score
12            Tone_z_score                 Tone score
13 submission_time_minutes Submission time in minutes
14 submission_time_minutes Submission time in minutes
15                      WC                 Word count
16                      WC                 Word count
17        Analytic_z_score             Analytic score
18        Analytic_z_score             Analytic score
19           Clout_z_score                Clout score
20           Clout_z_score                Clout score
21       Authentic_z_score            Authentic score
22       Authentic_z_score            Authentic score
23            Tone_z_score                 Tone score
24            Tone_z_score                 Tone score
25 submission_time_minutes Submission time in minutes
26 submission_time_minutes Submission time in minutes
27                      WC                 Word count
28                      WC                 Word count
29        Analytic_z_score             Analytic score
30        Analytic_z_score             Analytic score
31           Clout_z_score                Clout score
32           Clout_z_score                Clout score
33       Authentic_z_score            Authentic score
34       Authentic_z_score            Authentic score
35            Tone_z_score                 Tone score
36            Tone_z_score                 Tone score
37 submission_time_minutes Submission time in minutes
38 submission_time_minutes Submission time in minutes
39                      WC                 Word count
40                      WC                 Word count
41        Analytic_z_score             Analytic score
42        Analytic_z_score             Analytic score
43           Clout_z_score                Clout score
44           Clout_z_score                Clout score
45       Authentic_z_score            Authentic score
46       Authentic_z_score            Authentic score
47            Tone_z_score                 Tone score
48            Tone_z_score                 Tone score

我希望使用ifelse()命令更快地完成这项任务，并引用数据集，这就是我所称的快速方式。但是当我这样做的时候，它就不起作用了，产生了意想不到的结果。

我首先创建了一个变量，以消除列df_1$DV_name和df_2$Variable_analyses中的不可见字符。

### creates matching variables, which removes some invisible characters from data
# ---- NOTE: for df_1$DV_name, creating df_1$DV_name_for_matching
df_1$DV_name_for_matching <- 
  as.character(str_remove_all(df_1$DV_name, "[^A-z|0-9|[:punct:]|_|\\s]"))
# ---- NOTE: for df_2$Variable_analyses, creating 
df_2$Variable_analyses_for_matching <- 
  as.character(str_remove_all(df_2$Variable_analyses, "[^A-z|0-9|[:punct:]|_|\\s]"))

然后，我使用新变量df_1$DV_name_for_matching和df_2$Variable_analyses_for_matching作为ifelse()命令的基础：

### uses ifelse to complete matching task
df_1[["Variable_label"]] <- 
  ifelse(((df_1[["DV_name_for_matching"]]) == (df_2[["Variable_analyses_for_matching"]])), df_2[["Variable_label"]], NA)

这不会产生所需的输出(请参见上文)。相反，我得到了这个输出：

### displays df
# ---- NOTE: displays df, quick way does not work, not desired output
df_1

我不知道为什么快速方式不能工作。请告诉我如何使快速方式工作。

FYI，我在2013年英特尔Macbook Pro上使用RStudio。

谢谢。

下面是我用来创建帖子的代码

# creates df_1$Variable_label
# ---- NOTE: column(s) with values to be transfered - df_2$Variable_label
# ---- NOTE: column(s) for matching - df_1$DV_name, df_2$Variable_analyses

## displays data frames
df_1
df_2

## quick way
# ---- NOTE: quick way does not work

### creates matching variables, which removes some invisible characters from data
# ---- NOTE: for df_1$DV_name, creating df_1$DV_name_for_matching
df_1$DV_name_for_matching <- 
  as.character(str_remove_all(df_1$DV_name, "[^A-z|0-9|[:punct:]|_|\\s]"))
# ---- NOTE: for df_2$Variable_analyses, creating 
df_2$Variable_analyses_for_matching <- 
  as.character(str_remove_all(df_2$Variable_analyses, "[^A-z|0-9|[:punct:]|_|\\s]"))

### uses ifelse to complete matching task
df_1[["Variable_label"]] <- 
  ifelse(((df_1[["DV_name_for_matching"]]) == (df_2[["Variable_analyses_for_matching"]])), df_2[["Variable_label"]], NA)

### displays df
# ---- NOTE: displays df, quick way does not work, not desired output
df_1


## long way

### creates Variable_label
# ---- NOTE: does not directly extract Variable_label from df_2 and insert it into df_1
# ---- NOTE: based on df_1$Variable_label
df_1$Variable_label <- 
  ifelse(df_1$DV_name == "submission_time_minutes", "Submission time in minutes",
         ifelse(df_1$DV_name == "WC", "Word count",
                ifelse(df_1$DV_name == "Analytic_z_score", "Analytic score",
                       ifelse(df_1$DV_name == "Clout_z_score", "Clout score",
                              ifelse(df_1$DV_name == "Authentic_z_score", "Authentic score",
                                     ifelse(df_1$DV_name == "Tone_z_score", "Tone score", NA
                                     ))))))

### displays df
# ---- NOTE: displays df with created variable in desired output form
df_1

Answer 1

我相信你可以做一个left_join()。

library(tidyverse)

left_join(df_1, df_2, by = c("DV_name" = "Variable_analyses"))