Sort (order) data frame rows by multiple columns

user_84458 0
Viewed 1318181

I want to sort a data frame by multiple columns. For example, with the data frame below I would like to sort by column 'z' (descending) then by column 'b' (ascending):

dd <- data.frame(b = factor(c("Hi", "Med", "Hi", "Low"), 
      levels = c("Low", "Med", "Hi"), ordered = TRUE),
      x = c("A", "D", "A", "C"), y = c(8, 3, 9, 9),
      z = c(1, 1, 1, 2))
dd
    b x y z
1  Hi A 8 1
2 Med D 3 1
3  Hi A 9 1
4 Low C 9 2
r-faq dataframe r sorting
19 Answers

Suppose you have a data.frame A and you want to sort it using column called x descending order. Call the sorted data.frame newdata

newdata <- A[order(-A$x),]

If you want ascending order then replace "-" with nothing. You can have something like 

newdata <- A[order(-A$x, A$y, -A$z),]

where x and z are some columns in data.frame A. This means sort data.frame A by x descending, y ascending and z descending.

user_5890310

or you can use package doBy

library(doBy)
dd <- orderBy(~-z+b, data=dd)
user_1707920

if SQL comes naturally to you, sqldf package handles ORDER BY as Codd intended.

user_1467040

The arrange() in dplyr is my favorite option. Use the pipe operator and go from least important to most important aspect

dd1 <- dd %>%
    arrange(z) %>%
    arrange(desc(x))
user_104385130

Just for the sake of completeness, since not much has been said about sorting by column numbers... It can surely be argued that it is often not desirable (because the order of the columns could change, paving the way to errors), but in some specific situations (when for instance you need a quick job done and there is no such risk of columns changing orders), it might be the most sensible thing to do, especially when dealing with large numbers of columns.

In that case, do.call() comes to the rescue:

ind <- do.call(what = "order", args = iris[,c(5,1,2,3)])
iris[ind, ]

##        Sepal.Length Sepal.Width Petal.Length Petal.Width    Species
##    14           4.3         3.0          1.1         0.1     setosa
##    9            4.4         2.9          1.4         0.2     setosa
##    39           4.4         3.0          1.3         0.2     setosa
##    43           4.4         3.2          1.3         0.2     setosa
##    42           4.5         2.3          1.3         0.3     setosa
##    4            4.6         3.1          1.5         0.2     setosa
##    48           4.6         3.2          1.4         0.2     setosa
##    7            4.6         3.4          1.4         0.3     setosa
##    (...)
user_1891750

Another alternative, using the rgr package:

> library(rgr)
> gx.sort.df(dd, ~ -z+b)
    b x y z
4 Low C 9 2
2 Med D 3 1
1  Hi A 8 1
3  Hi A 9 1
user_11001070

I was struggling with the above solutions when I wanted to automate my ordering process for n columns, whose column names could be different each time. I found a super helpful function from the psych package to do this in a straightforward manner:

dfOrder(myDf, columnIndices)

where columnIndices are indices of one or more columns, in the order in which you want to sort them. More information here:

dfOrder function from 'psych' package

user_32768420
Related