Functions

SDS 192: Introduction to Data Science

Lindsay Poirier
Statistical & Data Sciences, Smith College

Fall 2022

For Today

• Quiz 2 Posted!
• Don’t forget about the Study Plan Assessment
• Writing Functions
• Code Along

Why write functions?

pit %>%
  filter(State == "AK" & Year == 2015) |>
  summarize(Total = sum(Count))

pit %>%
  filter(State == "AL" & Year == 2015) |>
  summarize(Total = sum(Count))

pit %>%
  filter(State == "AR" & Year == 2015) |>
  summarize(Total = sum(Count))

Why write functions?

• Reduces the amount of code to write
• Lowers the chances of errors in code
• Supports reproducibility

Functions

• Statements organized to perform a specific task
• Take arguments as inputs
  • Arguments can be required or optional
• Return a value or set of values as outputs
  • Default value is output of last line of function if not specified

User-defined Functions

• Basic format
• Example

function_name <- function(arg1, arg2) {
  
  x <- #Some code goes here referencing arg1 and arg2
  
  return(x)
}

calculate_difference <- function(arg1, arg2) {
  
  x <- arg1 - arg2
  
  return(x)
}

Calling Functions

• This is the same way you’d call functions built-in to R

• Basic format
• Example

function_name(value_for_arg1, value_for_arg2)

calculate_difference(7, 3)

[1] 4

User-defined Functions

• state_name is a required argument here.

calculate_state_total <- function(state_name) {
  
  x <- pit %>%
    filter(State == state_name & Year == 2015) |>
    summarize(Total = sum(Count))
  
  return(x)
}

calculate_state_total("AL")

Making Arguments Optional

• Because we provide a deafult value for year, it is optional in the function call.

• Setting Variable Defaults
• Overriding Variable Defaults

calculate_state_total <- function(state_name, year = 2020) {
  
  x <- pit %>%
    filter(State == state_name & Year == year) |>
    summarize(Total = sum(Count))
  
  return(x)
}

calculate_state_total("AL")

calculate_state_total <- function(state_name, year = 2020) {
  
  x <- pit %>%
    filter(State == state_name & Year == year) |>
    summarize(Total = sum(Count))
  
  return(x)
}

calculate_state_total("AL", 2019)

Naming Arguments

• Naming helps to differentiate between arguments. Order matters if arguments aren’t named!

calculate_state_total <- function(state_name, year = 2020) {
  
  x <- pit %>%
    filter(State == state_name & Year == year) |>
    summarize(Total = sum(Count))
  
  return(x)
}

calculate_state_total(state_name = "AL", year = 2019)