Functions & Control Flow

Writing reusable, intelligent R code

← Lab 3: dplyr Lab 4 of 10 Lab 5: Strings →

⏳ Loading R... (first load takes ~15 seconds)

📖 Concept Recap

Functions are the building blocks of reusable R code. Control flow lets your code make decisions and repeat actions.

Define functions: my_fn <- function(arg1, arg2 = default) { body }
if/else: if (condition) { ... } else { ... }
Vectorized if: ifelse(condition, yes, no)
for loops: for (item in vector) { ... }
while loops: while (condition) { ... }
sapply(): apply a function to each element of a vector, return a vector
lapply(): apply a function to each element, return a list

R returns the last expression in a function automatically — no explicit return() needed (though you can use it for early returns).

👀 Worked Example

A function with input validation, default arguments, and a named list return:

calculate_stats <- function(x, na.rm = TRUE) { if (!is.numeric(x)) stop("Input must be numeric") list( n = length(x), mean = mean(x, na.rm = na.rm), median = median(x, na.rm = na.rm), sd = sd(x, na.rm = na.rm), range = range(x, na.rm = na.rm) ) } scores <- c(88, 92, 75, 95, 84, 91, 78, NA) stats <- calculate_stats(scores) cat("N:", stats$n, "\n") cat("Mean:", round(stats$mean, 2), "\n") cat("Median:", stats$median, "\n") cat("SD:", round(stats$sd, 2), "\n")

✏️ Guided

Exercise 1 — Grade Converter

Complete the score_to_grade() function using if/else, then test it on a vector of scores with sapply().

score_to_grade <- function(score) {
  if (score >= 90) {
    "A"
  } else if (score >= 80) {
    "B"
  } else if (score >= 70) {
    "C"
  } else if (score >= 60) {
    "D"
  } else {
    "F"
  }
}

test_scores <- c(95, 83, 71, 65, 42, 88, 55, 90)

# Apply the function to every score at once
grades <- sapply(test_scores, score_to_grade)
cat("Scores:", test_scores, "\n")
cat("Grades:", grades, "\n")

# Count each grade
print(table(grades))

Output will appear here...

💡 Hint: sapply(vector, function_name) applies the function to each element and returns a vector of results. table() counts how many times each unique value appears.

💪 Independent

Exercise 2 — Vector Descriptor

Write a function describe_vector(x) that prints: N, mean, range (min–max), standard deviation, and count of outliers (values more than 2 SD from the mean). Test it on three different numeric vectors.

describe_vector <- function(x) {
  # Your function body here
  # Print: N, mean, range, SD, and outlier count
}

# Test on three vectors
v1 <- c(10, 12, 11, 13, 10, 45, 12, 11)   # has an outlier
v2 <- rnorm(20, mean = 100, sd = 15)
v3 <- c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10)

cat("=== Vector 1 ===\n");  describe_vector(v1)
cat("=== Vector 2 ===\n");  describe_vector(v2)
cat("=== Vector 3 ===\n");  describe_vector(v3)

Output will appear here...

💡 Hint: Outliers: sum(abs(x - mean(x)) > 2 * sd(x)). Use cat() to format output. Range: range(x) returns a 2-element vector — use range(x)[1] and range(x)[2] for min and max.

🔥 Challenge

Exercise 3 — FizzBuzz & Higher-Order Functions

Part A: Write fizzbuzz_r(n) that returns a character vector of length n: “FizzBuzz” for multiples of 15, “Fizz” for 3, “Buzz” for 5, otherwise the number as a string. Part B: Write run_multiple(fn, values) that applies fn to each element of values and prints the results. Use it to call fizzbuzz_r on c(10, 20, 30).

# Part A: FizzBuzz
fizzbuzz_r <- function(n) {
  # Return a character vector of length n
}

# Part B: Higher-order function
run_multiple <- function(fn, values) {
  # Apply fn to each value and print each result
}

# Test
cat("FizzBuzz(15):\n")
print(fizzbuzz_r(15))
cat("\nrun_multiple on c(10, 20, 30):\n")
run_multiple(fizzbuzz_r, c(10, 20, 30))

Output will appear here...

💡 Hint: Use ifelse() vectorized: ifelse(x %% 15 == 0, "FizzBuzz", ifelse(x %% 3 == 0, "Fizz", ...)). For run_multiple, use for (v in values) or lapply(values, fn).

🏆 Mini Project — Analysis Toolkit

Build a 4-Function Analysis Toolkit

Write and test all four functions below, then demonstrate them working together on a sample dataset.

clean_vector(x) — removes NAs and values > 3 SD from the mean, returns cleaned vector
compare_groups(data, group_col, value_col) — returns a data frame with mean, sd, and n per group
flag_outliers(x) — returns a logical vector: TRUE where value is > 2 SD from mean
full_report(x, label) — prints a formatted report using the other functions

clean_vector <- function(x) {
  # Remove NAs first
  x <- x[!is.na(x)]
  # Remove extreme outliers (> 3 SD from mean)
  x[abs(x - mean(x)) <= 3 * sd(x)]
}

compare_groups <- function(data, group_col, value_col) {
  # Return data frame with mean, sd, n per group
  groups <- unique(data[[group_col]])
  result <- data.frame(group=groups, mean=NA, sd=NA, n=NA)
  for (i in seq_along(groups)) {
    vals <- data[[value_col]][data[[group_col]] == groups[i]]
    result$mean[i] <- round(mean(vals), 2)
    result$sd[i]   <- round(sd(vals), 2)
    result$n[i]    <- length(vals)
  }
  result
}

flag_outliers <- function(x) {
  abs(x - mean(x, na.rm=TRUE)) > 2 * sd(x, na.rm=TRUE)
}

full_report <- function(x, label = "Data") {
  clean <- clean_vector(x)
  outlier_count <- sum(flag_outliers(x), na.rm=TRUE)
  cat("===", label, "Report ===\n")
  cat("Raw N:", length(x), " | Clean N:", length(clean), "\n")
  cat("Mean:", round(mean(clean), 2), " | SD:", round(sd(clean), 2), "\n")
  cat("Range:", min(clean), "to", max(clean), "\n")
  cat("Outliers flagged:", outlier_count, "\n\n")
}

# Test data
set.seed(1)
salaries <- c(rnorm(18, 60000, 10000), 200000, NA)
dept <- sample(c("Eng","Sales","HR"), 19, replace=TRUE)
df <- data.frame(dept=dept, salary=salaries[1:19])

full_report(salaries, "Salaries")
print(compare_groups(df, "dept", "salary"))

Output will appear here...

💡 Hint: Functions can call other functions you defined earlier in the same script. Build and test each function individually before combining them in full_report().

✅ Lab 4 Complete!

You can now write reusable, intelligent R functions with input validation, default arguments, control flow, and the apply family. These skills transform scripts into maintainable analysis tools.

Continue to Lab 5: String Manipulation →

← Lab 3: dplyr Lab 4 of 10 Lab 5: Strings →