CODEWITHSUNDEEP
rdplyrpivotreshapetidyr
wes
wes

Reputation: 113

Creating "Pivot Table" in R Populated with Factor Value for Each Row

I would like to take date input like this:

library(plyr)
names(ChickWeight) <- tolower(names(ChickWeight))
ChickWeight$diet.name <- as.factor(revalue(ChickWeight$diet, c("1" = "Grain", "2" = "Grubs", "3" = "Scraps", "4" = "Free Range")))

which creates:

> head(ChickWeight)
  weight time chick diet diet.name
1     42    0     1    1     Grain
2     51    2     1    1     Grain
3     59    4     1    1     Grain
4     64    6     1    1     Grain
5     76    8     1    1     Grain
6     93   10     1    1     Grain

> tail(ChickWeight)
    weight time chick diet  diet.name
573    155   12    50    4 Free Range
574    175   14    50    4 Free Range
575    205   16    50    4 Free Range
576    234   18    50    4 Free Range
577    264   20    50    4 Free Range
578    264   21    50    4 Free Range

And from those data, extract another set of data (chick.feed) that looks like this:

chick   diet.name
    1       Grain
    2       Grain
    .           .
    .           .
    .           .
   21       Grubs
   22       Grubs
    .           .
    .           .
    .           .
   31      Scraps
   32      Scraps
    .           .
    .           .
    .           .
   49  Free Range
   50  Free Range

I.e., I would like to create a "pivot table" (chick.feed) in which the computed value for each row (here, each chick) is the alphabetic value ("factor" in R-speak) of all the various rows for that chick in the ChickWeight.

(My question appears to be similar but not identical to this one about converting from long to wide data.)

Upvotes: 0

Views: 142

Answers (1)

DanY
DanY

Reputation: 6073

No need to reshape. Just grab the desired columns and keep the unique rows:

unique(ChickWeight[ , c("chick", "diet.name")])

Upvotes: 1

Related Questions