Sorting

Problem

You want to sort a vector, matrix, or data frame.

Solution

Vectors

  1. # Make up a randomly ordered vector
  2. v <- sample(101:110)
  4. # Sort the vector
  5. sort(v)
  6. #>  [1] 101 102 103 104 105 106 107 108 109 110
  8. # Reverse sort
  9. sort(v, decreasing=TRUE)
  10. #>  [1] 110 109 108 107 106 105 104 103 102 101

Data frames

To sort a data frame on one or more columns, you can use the arrange function from plyr package, or use R’s built-in functions. The arrange function is much easier to use, but does require the external package to be installed.

  1. # Make a data frame
  2. df <- data.frame (id=1:4,
  3.             weight=c(20,27,24,22),
  4.             size=c("small", "large", "medium", "large"))
  5. df
  7. library(plyr)
  9. # Sort by weight column. These have the same result.
  10. arrange(df, weight)       # Use arrange from plyr package
  11. df[ order(df$weight), ]   # Use built-in R functions
  12. #>   id weight   size
  13. #> 1  1     20  small
  14. #> 2  4     22  large
  15. #> 3  3     24 medium
  16. #> 4  2     27  large
  19. # Sort by size, then by weight
  20. arrange(df, size, weight)         # Use arrange from plyr package
  21. df[ order(df$size, df$weight), ]  # Use built-in R functions
  22. #>   id weight   size
  23. #> 4  4     22  large
  24. #> 2  2     27  large
  25. #> 3  3     24 medium
  26. #> 1  1     20  small
  29. # Sort by all columns in the data frame, from left to right
  30. df[ do.call(order, as.list(df)), ] 
  31. # In this particular example, the order will be unchanged

Note that the size column is a factor and is sorted by the order of the factor levels. In this case, the levels were automatically assigned alphabetically (when creating the data frame), so large is first and small is last.

Reverse sort

The overall order of the sort can be reversed with the argument decreasing=TRUE.

To reverse the direction of a particular column, the method depends on the data type:

  • Numbers: put a - in front of the variable name, e.g. df[ order(-df$weight), ].
  • Factors: convert to integer and put a - in front of the variable name, e.g. df[ order(-xtfrm(df$size)), ].
  • Characters: there isn’t a simple way to do this. One method is to convert to a factor first and then sort as above.
  1. # Reverse sort by weight column. These all have the same effect:
  2. arrange(df, -weight)                      # Use arrange from plyr package
  3. df[ order(df$weight, decreasing=TRUE), ]  # Use built-in R functions
  4. df[ order(-df$weight), ]                  # Use built-in R functions
  5. #>   id weight   size
  6. #> 2  2     27  large
  7. #> 3  3     24 medium
  8. #> 4  4     22  large
  9. #> 1  1     20  small
  12. # Sort by size (increasing), then by weight (decreasing)
  13. arrange(df, size, -weight)         # Use arrange from plyr package
  14. df[ order(df$size, -df$weight), ]  # Use built-in R functions
  15. #>   id weight   size
  16. #> 2  2     27  large
  17. #> 4  4     22  large
  18. #> 3  3     24 medium
  19. #> 1  1     20  small
  22. # Sort by size (decreasing), then by weight (increasing)
  23. # The call to xtfrm() is needed for factors
  24. arrange(df, -xtfrm(size), weight)         # Use arrange from plyr package
  25. df[ order(-xtfrm(df$size), df$weight), ]  # Use built-in R functions
  26. #>   id weight   size
  27. #> 1  1     20  small
  28. #> 3  3     24 medium
  29. #> 4  4     22  large
  30. #> 2  2     27  large