CODEWITHSUNDEEP
rtidyrspread
E Norton
E Norton

Reputation: 95

Spreading my dataframe with multiple unique identifiers

I'm trying to create a dataframe that spreads my dataframe from this:

my_data1 <- data.frame(Tank = c("1A", "1D", "1F", "1A", "1F", "2A"), Bag = c("C1", "C2", "C2", "C1", "C2", "T1"), name = c("ally", "ally", "nancy", "ted", "mark", "ted"), deformity=c("spinal", "eye", "snout", "spinal", "head", "eye"))

to this:

my_data2 <- data.frame(Tank = c("1A", "1D", "1F", "2A"), Bag = c("C1", "C2", "C2", "T1"), ally_deformity = c("spinal", "eye", NA, NA), nancy_deformity=c(NA, NA, "snout", NA), ted_deformity=c("spinal", NA, NA, "eye"), mark_deformity=c(NA, NA, "head", NA))

So I want the columns "Tank" and "Bag" to be my unique identifiers and every other column to be the deformities each individual person found for that specific tank and bag. I've been trying tidyr's spread with no luck.

Upvotes: 1

Views: 55

Answers (3)

ThomasIsCoding
ThomasIsCoding

Reputation: 101343

A data.table option using dcast

dcast(setDT(my_data1), Tank + Bag ~ paste0(name, "_deformity"))

gives

   Tank Bag ally_deformity mark_deformity nancy_deformity ted_deformity
1:   1A  C1         spinal           <NA>            <NA>        spinal
2:   1D  C2            eye           <NA>            <NA>          <NA>
3:   1F  C2           <NA>           head           snout          <NA>
4:   2A  T1           <NA>           <NA>            <NA>           eye

Upvotes: 0

Onyambu
Onyambu

Reputation: 79228

Using base R, you could do:

reshape(my_data1, v.names = "deformity", timevar = "name", dir="wide", idvar = "Tank", sep="_")

  Tank Bag deformity_ally deformity_nancy deformity_ted deformity_mark
1   1A  C1         spinal            <NA>        spinal           <NA>
2   1D  C2            eye            <NA>          <NA>           <NA>
3   1F  C2           <NA>           snout          <NA>           head
6   2A  T1           <NA>            <NA>           eye           <NA>

Upvotes: 1

akrun
akrun

Reputation: 887118

We can use pivot_wider

library(dplyr)
library(tidyr)
library(stringr)
my_data1 %>%
  mutate(name = str_c(name, '_deformity')) %>%
  pivot_wider(names_from = name, values_from = deformity)

-output

# A tibble: 4 x 6
#  Tank  Bag   ally_deformity nancy_deformity ted_deformity mark_deformity
#  <chr> <chr> <chr>          <chr>           <chr>         <chr>         
#1 1A    C1    spinal         NA              spinal        NA            
#2 1D    C2    eye            NA              NA            NA            
#3 1F    C2    NA             snout           NA            head          
#4 2A    T1    NA             NA              eye           NA

Upvotes: 1

Related Questions