Search code examples
rdataframedplyrgroup-byaggregate

Make a list from dataframe column

From a data frame df (a and b are column names)

a b
xx Apple
yy Orange
zz Apple
dd Mango
pp Mango

I would like the output as

Apple xx,zz
Orange yy
Mango dd,pp

I tried aggregate and group_by but failed.

Solution

  • Base R:

    A single aggregate would be enough for this operation. Here you would apply FUN to column a with a grouping on column b (a ~ b). The function to use is paste with argument collapse so that the multiple strings would be collapsed into a single one.

    aggregate(a ~ b, df, FUN = paste, collapse = ",")

       b     a
1  Apple xx,zz
2  Mango dd,pp
3 Orange    yy

    Dplyr

    Since you mentioned group_by, the correct syntax to do so in dplyr is as follows:

    library(dplyr)

df %>% group_by(b) %>% summarize(a = paste(a, collapse = ","))

# A tibble: 3 × 2
  b      a    
  <chr>  <chr>
1 Apple  xx,zz
2 Mango  dd,pp
3 Orange yy