Add column by ID into another df with duplicate IDs

[bayesian]

I would like to join the column Product from my original dataset into another df with tokenised words based on Complaint.ID. I am not sure how to go about it because there are duplicate IDs in df2.

## Reprex
# original dataset
df1 <- tibble::tribble(
      ~Complaint.ID,        ~Product,
                  1,         "Credit card",
                  2,         "Student loan",
                  3,         "Mortgage",
                  4,         "Car Insurance",
                  5,         "Credit card",
                  6,         "Mortgage",
    
    )

# dataset of tokenised words
df2 <- tribble(
  ~Complaint.ID,          ~word,
            1,         "come",
            1,    "attention",
            1,         "citi",
            2,        "group",
            2,     "actively",
            3,   "attempting",
            3,    "interfere",
            3,       "rights",
            3,   "guaranteed",
            3, "constitution"
)

> df1
# A tibble: 6 x 2
  Complaint.ID Product      
         <dbl> <chr>        
1            1 Credit card  
2            2 Student loan 
3            3 Mortgage     
4            4 Car Insurance
5            5 Credit card  
6            6 Mortgage     
> df2
# A tibble: 10 x 2
   Complaint.ID word        
          <dbl> <chr>       
 1            1 come        
 2            1 attention   
 3            1 citi        
 4            2 group       
 5            2 actively    
 6            3 attempting  
 7            3 interfere   
 8            3 rights      
 9            3 guaranteed  
10            3 constitution

This should be what it looks like in the end:

> joined_df
# A tibble: 10 x 3
   Complaint.ID word         Product        
          <dbl> <chr>        <chr>
 1            1 come         Credit card
 2            1 attention    Credit card
 3            1 citi         Credit card
 4            2 group        Student loan
 5            2 actively     Student loan
 6            3 attempting   Mortgage
 7            3 interfere    Mortgage
 8            3 rights       Mortgage
 9            3 guaranteed   Mortgage
10            3 constitution Mortgage

I've tried out the different dplyr joins but they are not working for me. Thanks for any help!

[StatSteph]

I'm not sure what you tried but I'd suggest using left_join as follows:

library(tidyverse)

df1 <- tibble::tribble(
  ~Complaint.ID,        ~Product,
  1,         "Credit card",
  2,         "Student loan",
  3,         "Mortgage",
  4,         "Car Insurance",
  5,         "Credit card",
  6,         "Mortgage",
  
)

# dataset of tokenised words
df2 <- tribble(
  ~Complaint.ID,          ~word,
  1,         "come",
  1,    "attention",
  1,         "citi",
  2,        "group",
  2,     "actively",
  3,   "attempting",
  3,    "interfere",
  3,       "rights",
  3,   "guaranteed",
  3, "constitution"
)

joined_df <- df2 %>% left_join(df1, by="Complaint.ID")
joined_df
#> # A tibble: 10 x 3
#>    Complaint.ID word         Product     
#>           <dbl> <chr>        <chr>       
#>  1            1 come         Credit card 
#>  2            1 attention    Credit card 
#>  3            1 citi         Credit card 
#>  4            2 group        Student loan
#>  5            2 actively     Student loan
#>  6            3 attempting   Mortgage    
#>  7            3 interfere    Mortgage    
#>  8            3 rights       Mortgage    
#>  9            3 guaranteed   Mortgage    
#> 10            3 constitution Mortgage

^{Created on 2020-12-15 by the reprex package (v0.3.0)}

[bayesian]

Somehow I completely overlooked this. Thanks!