如何用dplyr更简便编码？求值转二进制/字符的快速方法

Hey there! Let's tackle your two questions one by one, using tidyverse-friendly approaches and some handy R tricks:

Your original base R code works, but we can streamline it entirely within a dplyr workflow—no need to create a separate vector and modify it. Here are a few clean options:

Option 1: Use if_else() + factor() (most straightforward for binary cases) #

library(dplyr)

college <- college %>%
  mutate(elite = factor(if_else(Top10perc > 50, "Yes", "No")))

Option 2: Use case_when() (great if you might add more conditions later) #

Even for a binary check, case_when() keeps your code scalable:

college <- college %>%
  mutate(elite = factor(case_when(
    Top10perc > 50 ~ "Yes",
    TRUE ~ "No"  # Catch-all for all other cases
  )))

Option 3: Ultra-minimal logical-to-factor conversion #

Since Top10perc > 50 returns a logical vector, we can directly convert it to a factor with custom labels:

college <- college %>%
  mutate(elite = factor(Top10perc > 50, 
                        levels = c(FALSE, TRUE), 
                        labels = c("No", "Yes")))

All these methods keep your code chained to the data frame, making it easier to follow and extend.

Binary (0/1) Values #

If you want a numeric binary column instead of a factor/character, use these fast approaches:

• Leverage R's automatic logical-to-integer conversion: as.integer(Top10perc > 50) (TRUE becomes 1, FALSE becomes 0)
• Explicit assignment with if_else(): if_else(Top10perc > 50, 1L, 0L) (the L ensures integer type instead of numeric)
• Use dplyr::recode() for clarity: recode(Top10perc > 50, TRUE= 1L,FALSE = 0L)

Character Values #

To get a character vector instead of a factor, just remove the factor() wrapper from the examples above:

• dplyr approach: mutate(elite = if_else(Top10perc > 50, "Yes", "No"))
• Base R shortcut: elite <- ifelse(Top10perc > 50, "Yes", "No")

A bonus tip: For easy conversions between factor, character, and binary later, use the forcats package (part of the tidyverse). For example:

• Convert factor to character: as.character(elite)
• Convert factor to binary (match labels to 0/1): as.integer(elite) - 1 (adjust based on your factor levels)

内容的提问来源于stack exchange，提问作者Bury