Using a function to add a column in R data.frame

Question

I am trying to write a function in R which lumps species columns together within a data.frame.

(To elaborate a bit on what I'm doing...I have a data frame with multiple plant species for multiple sites and years. Some of the species were misidentified, so I'd like to group to a more general level (e.g. spp a and spp b were mixed up throughout the years; so I'd like to create a new column called spp.ab in which the data for spp a and b are lumped together)).

Example:

spp.a spp.b
  1     0
  2     3
  0     4
  3     2
  4     5

I'd like to eventually end up with a single column that displays the maximum from value from the two species:

spp.ab
  1
  3
  4
  3
  5

I've started writing a function which does this; however, I'm having troubling adding the new column to my data set and dropping the old ones. Could someone tell me what's wrong with my code?

lump <- function(db, spp.list, new.spp) { #input spp.list as c('spp.a', 'spp.b', ...)
  mini.db <- subset(db, select=spp.list);
  newcol <- as.vector(apply(mini.db, 1, max, na.rm=T));
  db$new.spp <- newcol
  db <- db[,names(db) %in% spp.list]
  return(db)
}

When I call the function as such

test <- lump(db, c('spp.a', 'spp.b'), spp.ab)
test

all that pops up is the mini.db. Am I missing something with return()?

For reference, db is the database, spp.list is the species I want to lump together, and new.spp is what I would like the new column named.

Thanks for any help,
Paul

logicForPresident · Accepted Answer · 2013-12-21 00:49:34Z

3

I've figured it out...stupid mistake, of course. Here is the code that works:

lump <- function(db, spp.list, new.spp) { #input spp.list as a c('spp.a', 'spp.b', ...), and new.spp must be in quotes (e.g. 'new.spp')
    mini.db <- subset(db, select=spp.list);
    newcol <- as.vector(apply(mini.db, 1, max, na.rm=T));
    newcol[newcol==-Inf] <- NA;
    db[new.spp] <- newcol;
    db <- db[, !names(db) %in% spp.list];
    return(as.data.frame(db));
 }

The key is in the db[new.spp] <- newcol; line. Apparently using this works, but using db$new.spp <- newcol does not. I also then added a ! to the line db <- db[,!names(db) %in% spp.list]. This was my biggest mistake.

answered Dec 21, 2013 at 0:49

logicForPresident

3116 silver badges20 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

Julián Urbano Over a year ago

So new.spp is the name of the new column?

logicForPresident Over a year ago

Correct, and you can call it whatever you like by placing the name in quotes as an argument to the function. For example, if you want to call it el.conquistador you would input lump(db, c('spp.a', 'spp.b'), 'el.conquistador')

A5C1D2H2I1M1N2O1R2T1 · Accepted Answer · 2013-12-21 01:26:59Z

2

While it seems like you've found your answer, I would suggest, instead, the pmax function:

> with(db, pmax(spp.a, spp.b))
[1] 1 3 4 3 5

You can use this with within or transform to mimic your function:

out <- within(db, spp.ab <- pmax(spp.a, spp.b))
out
#   spp.a spp.b spp.ab
# 1     1     0      1
# 2     2     3      3
# 3     0     4      4
# 4     3     2      3
# 5     4     5      5

edited Dec 21, 2013 at 1:26

answered Dec 21, 2013 at 1:18

A5C1D2H2I1M1N2O1R2T1

194k31 gold badges417 silver badges497 bronze badges

Collectives™ on Stack Overflow

Using a function to add a column in R data.frame

2 Answers 2

2 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related