CODEWITHSUNDEEP
r
wenge
wenge

Reputation: 59

R query on '$'()

I created a data frame and am trying to access the columns in the data frame. The code is as shown below

df <- data.frame(n = c('A','B','C'),
                 mark_1 = c(23,25, 17),
                 mark_2 = c(45,46,50))

j <- paste('mark',1,sep="_")
j

f <- '$'(df,j)
summary(f)

n <- '$'(df,"mark_1")
summary(n)

When I run this code, this is what I get

    > j <- paste('mark',1,sep="_")
    > j
    [1] "mark_1"
    > f <- '$'(df,j)
    > summary(f)
    Length  Class   Mode 
         0   NULL   NULL 
    > n <- '$'(df,"mark_1")
    > summary(n)
       Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
      17.00   20.00   23.00   21.67   24.00   25.00 
    >

Why is summary(f) not working? j is properly concatenated to "mark_1". Then why is the error happening?

Thanks in advance

Upvotes: 2

Views: 156

Answers (4)

IRTFM
IRTFM

Reputation: 263481

You could use '$' if you applied it with do.call. do.call is useful in situations where you have the name of the function as a character rather than a language value. In this case it also evaluates the second argument, the list of dataframe and column name.

do.call('$', list(df, j))
[1] 23 25 17

That said, I think you are better off using '[[' or '[' which do the evaluation of the second argument as a matter of course.

Upvotes: 2

Tommy
Tommy

Reputation: 40871

@TylerRinker showed you what to do instead.

...but the reason you can't use '$'(df,j) is that the dollar operator uses the second argument literally.

...so '$'(df,mark_1) would work fine. But you simply can't pass a variable to it. That's what [ and [[ are for:

df[,j]  # The j'th column
df[[j]] # Pretend df is a list of columns and get the j'th element

Upvotes: 6

Vincent Zoonekynd
Vincent Zoonekynd

Reputation: 32401

'$'(df,j) is equivalent to df$j -- it looks at the name of the argument, rather than its value. You can use df[[j]] or df[,j] or '[['(df,j) instead (but avoid the last one, unless you want to make the code difficult to read).

df$j <- 1

# The "j" column
df$j
`$`(df,j)
`[[`(df,"j")
df[["j"]]
df[,"j"]

# The "mark_1" column
df[[j]]
df[,j]
`[[`(df,j)

Upvotes: 4

Tyler Rinker
Tyler Rinker

Reputation: 110062

Use indexing over the $ as in:

df[, j]
summary(df[, j])

Upvotes: 5

Related Questions