How to Find Mode of a Vector in R

1. Use Custom Function to Find Mode of an R Vector
2. Use map_dbl to Apply FindMode Function to Each Data Frame Column in R

This article will explain several methods of how to find a mode of a vector in R.

Use Custom Function to Find Mode of an R Vector

Mode is one of the most basic statistical concepts denoting the maximum occurrence value in a set of values. As a result, it can be observed in different data types, such as numeric or character-based. R language does not have a built-in function for calculating the mode, but we can implement it using the functions: unique, which.max, tabulate and match. We define a function named FindMode that takes one argument denoted as x. At first, the unique function is called on x and stored in a separate object. unique extracts non-duplicate values from the set. It can take a vector object, data frame, or an array as the first argument.

Then, we have multiple chained functions starting with match that takes two vectors and returns a vector of position matches from them. tabulate function counts and returns the number of times each integer appears in a vector. Note that the returned vector contains a count for every integer that is less or equal to the maximum integer in the argument vector. Finally, the which.max function finds the index of the maximum element in the integer vector. We can invoke the FindMode function on every column of the data frame using the apply function. In this case, we declare a simple integer vector stored as a data frame and then passed to the apply function to calculate the mean.

library(purrr)

FindMode <- function(x) {
  u <- unique(x)
  u[which.max(tabulate(match(x, u)))]
}

x <- c(12, 44, 3, -4.2, 3, 3.2, 54, 4, -11, -8, 2.5)
df <- data.frame(x)
apply(df, 2, FindMode)

Output:

x
3

Alternatively, we can call the FindMode function on the data frame directly. The following example code demonstrates the usage with the cars dataset included in the datasets package.

library(purrr)

FindMode <- function(x) {
  u <- unique(x)
  u[which.max(tabulate(match(x, u)))]
}

apply(cars, 2, FindMode)

Output:

speed  dist
   20    26

Use map_dbl to Apply FindMode Function to Each Data Frame Column in R

Another useful function to find the mean for each column of the given data frame is map_dbl, which is part of the purrr package included in the tidyverse. Note that these methods are not to be called on the vector objects.

library(purrr)

FindMode <- function(x) {
  u <- unique(x)
  u[which.max(tabulate(match(x, u)))]
}
x <- c(12, 44, 3, -4.2, 3, 3.2, 54, 4, -11, -8, 2.5)
df <- data.frame(x)

map_dbl(df, FindMode)
map_dbl(cars, FindMode)

Output:

x
3
speed  dist
   20    26