Lab 4: Vectors & Missing Values

BI 255 · Bethune-Cookman University · Fall 2026

Author

Dr. Rosie Stanbrook-Buyer

Published

August 26, 2026

Vectors & Missing Data in the NFL

Real data is messy. Players get injured and miss games. Stats aren’t always recorded. Measurement errors happen. In the NFL, a quarterback might play 14 of 17 games — the other 3 are NA (Not Available) in your dataset.

Today we learn how to build vectors properly and how to handle the inevitable missing values that appear in real biological and sports data.

Learning Objectives

By the end of this lab you will be able to:

Build and name vectors using c() and names()
Understand what NA means and why it matters
Use is.na() to detect missing values
Use na.rm = TRUE to calculate statistics despite missing data
Replace or remove NA values
Perform vector arithmetic

Part 1: Building Vectors Properly

A vector is R’s most fundamental data structure — a sequence of values all of the same type.

# Patrick Mahomes' passing yards per game, 2024 season (first 10 games)
mahomes_yards <- c(267, 331, 198, 312, 279, 290, 341, 188, 255, 320)
mahomes_yards

 [1] 267 331 198 312 279 290 341 188 255 320

# Basic information about the vector
length(mahomes_yards)   # How many elements?

[1] 10

class(mahomes_yards)    # What type?

[1] "numeric"

Named Vectors

You can label each element in a vector, which makes your data much more readable:

# Give each value a name corresponding to the opponent
opponents <- c("BAL", "CIN", "ATL", "LAC", "NO",
               "LV", "TB", "DEN", "CAR", "LV")

names(mahomes_yards) <- opponents
mahomes_yards

BAL CIN ATL LAC  NO  LV  TB DEN CAR  LV 
267 331 198 312 279 290 341 188 255 320

# Now you can see which game each number belongs to
# Access by name
mahomes_yards["BAL"]

BAL 
267

mahomes_yards["TB"]

 TB 
341

Why Name Your Vectors?

Names make vectors much easier to read and interpret. When you come back to your code weeks later, mahomes_yards["BAL"] is much clearer than mahomes_yards[1].

Part 2: Missing Values — `NA`

NA stands for Not Available. It is R’s placeholder for missing data.

Think of it like a blank cell in a spreadsheet — the data should be there, but it isn’t.

# Lamar Jackson's rushing yards per game, 2024 season
# He missed 2 games due to injury (games 6 and 11)
jackson_rushing <- c(45, 62, 38, 71, 55, NA, 49, 83, 60, 41, NA, 77)
jackson_rushing

 [1] 45 62 38 71 55 NA 49 83 60 41 NA 77

# R tells you there are 2 NAs
summary(jackson_rushing)

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max.     NAs 
  38.00   46.00   57.50   58.10   68.75   83.00       2

Detecting Missing Values with `is.na()`

# Which positions are NA?
is.na(jackson_rushing)

 [1] FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE

# How many games did he miss?
sum(is.na(jackson_rushing))

[1] 2

# How many games did he play?
sum(!is.na(jackson_rushing))   # ! means "NOT"

[1] 10

️ NAs are Contagious!

If any value in your calculation is NA, the result is also NA:

mean(jackson_rushing)   # Returns NA — because of the missing games!

[1] NA

This seems annoying, but it’s actually protecting you. R is saying: “I don’t know the answer because some data is missing.”

Part 3: Working Around Missing Values

`na.rm = TRUE`

Most R statistical functions have an na.rm argument. Setting it to TRUE tells R to remove the NAs before calculating.

# Now we get a real answer
mean(jackson_rushing, na.rm = TRUE)

[1] 58.1

sum(jackson_rushing,  na.rm = TRUE)

[1] 581

max(jackson_rushing,  na.rm = TRUE)

[1] 83

min(jackson_rushing,  na.rm = TRUE)

[1] 38

What does na.rm stand for?

na.rm = NA remove. Setting it to TRUE removes NAs before doing the calculation.

Removing NAs Entirely

Sometimes you want a clean vector with no NAs at all:

# na.omit() removes all NA values
jackson_clean <- na.omit(jackson_rushing)
jackson_clean

 [1] 45 62 38 71 55 49 83 60 41 77
attr(,"na.action")
[1]  6 11
attr(,"class")
[1] "omit"

length(jackson_clean)   # 10 games (the 2 missed are gone)

[1] 10

# Alternatively, use logical indexing (we'll cover this more in Lab 6)
jackson_played <- jackson_rushing[!is.na(jackson_rushing)]
jackson_played

 [1] 45 62 38 71 55 49 83 60 41 77

Replacing NAs with a Value

Sometimes you want to replace NA with something meaningful (like 0 for a missed game):

jackson_with_zeros <- jackson_rushing
jackson_with_zeros[is.na(jackson_with_zeros)] <- 0
jackson_with_zeros

 [1] 45 62 38 71 55  0 49 83 60 41  0 77

Should You Replace NAs with Zero?

It depends on the context! For rushing yards, replacing with 0 makes sense (if he didn’t play, he gained 0 yards). But for body temperature or a test score, replacing missing data with 0 would be wrong and misleading.

In science, always document how you handled missing values and justify your choice.

Part 4: Vector Arithmetic

Like we saw with sequences, arithmetic on vectors applies to every element:

# Points scored by the Dallas Cowboys in each week of the season
cowboys_pts <- c(24, 17, 31, 20, 28, 14, 35, 22, 19, 27,
                 30, 21, 16, 28, 33, 18, 24)

# Points scored by opponents (what they allowed)
cowboys_allowed <- c(20, 21, 28, 17, 31, 20, 28, 30, 22, 14,
                     17, 24, 27, 21, 30, 27, 20)

# Point differential for each game (positive = win, negative = loss)
point_diff <- cowboys_pts - cowboys_allowed
point_diff

 [1]   4  -4   3   3  -3  -6   7  -8  -3  13  13  -3 -11   7   3  -9   4

# Which games did they win?
won_game <- point_diff > 0
won_game

 [1]  TRUE FALSE  TRUE  TRUE FALSE FALSE  TRUE FALSE FALSE  TRUE  TRUE FALSE
[13] FALSE  TRUE  TRUE FALSE  TRUE

# How many wins?
sum(won_game)

[1] 9

# How many losses?
sum(point_diff < 0)

[1] 8

# Total points scored this season
sum(cowboys_pts)

[1] 407

# Average points per game
round(mean(cowboys_pts), 1)

[1] 23.9

Arithmetic Between Two Vectors

# Suppose each point is worth different fantasy values
# 1 point = 1 fantasy point, but bonus points for games over 28

bonus <- ifelse(cowboys_pts > 28, 3, 0)   # 3 bonus points for high-scoring games
fantasy_score <- cowboys_pts + bonus
fantasy_score

 [1] 24 17 34 20 28 14 38 22 19 27 33 21 16 28 36 18 24

ifelse() — Your First Conditional Function

ifelse(test, value_if_TRUE, value_if_FALSE) applies a test to every element and returns one of two values.

This is extremely useful and you’ll use it constantly when cleaning data.

Part 5: Putting It Together — QB Comparison

Let’s compare two quarterbacks across a shared 10-game stretch, handling missing data properly:

# 10-game passer rating comparison
qb_names <- c("Game 1", "Game 2", "Game 3", "Game 4", "Game 5",
               "Game 6", "Game 7", "Game 8", "Game 9", "Game 10")

# Patrick Mahomes passer ratings (0-158.3 scale)
mahomes_rating <- c(112.4, 98.7, 131.2, NA, 105.8,
                    122.1, 89.3, 141.5, 107.6, 118.9)

# Jalen Hurts passer ratings
hurts_rating   <- c(98.1, 107.5, 88.4, 125.3, 110.2,
                    NA, 119.7, 92.8, NA, 130.1)

# How many games did each miss?
cat("Mahomes missed:", sum(is.na(mahomes_rating)), "game(s)\n")

Mahomes missed: 1 game(s)

cat("Hurts missed:  ", sum(is.na(hurts_rating)), "game(s)\n")

Hurts missed:   2 game(s)

# Average passer rating (ignoring missed games)
mahomes_avg <- mean(mahomes_rating, na.rm = TRUE)
hurts_avg   <- mean(hurts_rating,   na.rm = TRUE)

cat("Mahomes average passer rating:", round(mahomes_avg, 1), "\n")

Mahomes average passer rating: 114.2

cat("Hurts average passer rating:  ", round(hurts_avg,   1), "\n")

Hurts average passer rating:   109

cat("Difference:", round(mahomes_avg - hurts_avg, 1), "points\n")

Difference: 5.2 points

# Simple side-by-side bar chart
barplot(
  c(mahomes_avg, hurts_avg),
  names.arg = c("P. Mahomes", "J. Hurts"),
  col       = c("#E31837", "#004C54"),   # Chiefs red, Eagles midnight green
  ylim      = c(0, 160),
  ylab      = "Average Passer Rating",
  main      = "QB Passer Rating Comparison\n(missing games excluded)",
  border    = "white"
)
abline(h = 100, col = "gray50", lty = 2)
text(0.7, 102, "100 = Good QB Rating", col = "gray30", cex = 0.8)

3-Minute Knowledge Check

Close your notes. Answer these on your own — you have 3 minutes. We’ll go through the answers together.

Knowledge Check Questions

1. What does NA stand for in R?

a) Not Applicable  &nbsp;&nbsp; b) Not Available  &nbsp;&nbsp; c) Null Answer  &nbsp;&nbsp; d) Negative Amount

2. You run mean(c(10, 20, NA, 40)) and get NA. How do you fix this?

3. What does sum(is.na(x)) calculate?

a) The sum of all values in x  &nbsp;&nbsp; b) The number of missing values in x  &nbsp;&nbsp; c) Whether x has any NAs  &nbsp;&nbsp; d) The mean of x

4. True or False: na.omit(x) replaces NAs with zero.

5. You have a vector scores <- c(88, NA, 91, 75, NA). Write the code to calculate the mean ignoring the missing values.

Answers (Only reveal if you have answered the above questions)

1. b) Not Available — NA is R’s way of representing missing or unknown data.

2. Add na.rm = TRUE: mean(c(10, 20, NA, 40), na.rm = TRUE) → returns 23.33. The na.rm argument tells R to remove NAs before calculating.

3. b) The number of missing values in x — is.na(x) returns a logical vector of TRUE/FALSE, and sum() counts the TRUEs (each TRUE = 1).

4. False — na.omit(x) removes the NA values entirely, returning a shorter vector. To replace NAs with zero, you would use x[is.na(x)] <- 0.

5. mean(scores, na.rm = TRUE) → returns 84.67 (average of 88, 91, and 75).

Lab 4 Checklist

Before you leave, make sure you can:

Create a named vector using c() and names()
Explain what NA means in R
Use is.na() to find missing values
Use sum(is.na()) to count missing values
Calculate mean(), sum(), etc. with na.rm = TRUE
Use na.omit() to remove NAs
Replace NAs using logical indexing x[is.na(x)] <- value
Perform arithmetic between two vectors

Bonus Challenge

Find the season statistics for two NFL quarterbacks (any two you like). Create named vectors for their touchdown passes in each week. One of them should have at least one NA for a missed game. Compare their averages using na.rm = TRUE and make a barplot. Add this to your Quarto document and render it.

Before Next Class (Friday)

Read Intro2r §3.1, 3.2
Friday (Lab 5) we move to data frames and matrices — where rows and columns begin
Practice: create a vector with at least one NA and calculate its mean both with and without na.rm = TRUE

--- title: "Lab 4: Vectors & Missing Values" subtitle: "BI 255 · Bethune-Cookman University · Fall 2026" author: "Dr. Rosie Stanbrook-Buyer" date: "August 26, 2026" format: html: theme: cosmo toc: true toc-location: left toc-title: "In This Lab" toc-depth: 3 code-fold: false code-tools: true highlight-style: github smooth-scroll: true embed-resources: true callout-appearance: default execute: warning: false message: false echo: true --- ```{r setup, include=FALSE} knitr::opts_chunk$set(echo = TRUE) ``` ------------------------------------------------------------------------ ## Vectors & Missing Data in the NFL Real data is messy. Players get injured and miss games. Stats aren't always recorded. Measurement errors happen. In the NFL, a quarterback might play 14 of 17 games — the other 3 are `NA` (Not Available) in your dataset. Today we learn how to build **vectors** properly and how to handle the inevitable **missing values** that appear in real biological and sports data. ::: callout-note ## Learning Objectives By the end of this lab you will be able to: 1. Build and name vectors using `c()` and `names()` 2. Understand what `NA` means and why it matters 3. Use `is.na()` to detect missing values 4. Use `na.rm = TRUE` to calculate statistics despite missing data 5. Replace or remove `NA` values 6. Perform vector arithmetic ::: ------------------------------------------------------------------------ ## Part 1: Building Vectors Properly A **vector** is R's most fundamental data structure — a sequence of values all of the **same type**. ```{r} # Patrick Mahomes' passing yards per game, 2024 season (first 10 games) mahomes_yards <- c(267, 331, 198, 312, 279, 290, 341, 188, 255, 320) mahomes_yards ``` ```{r} # Basic information about the vector length(mahomes_yards) # How many elements? class(mahomes_yards) # What type? ``` ### Named Vectors You can label each element in a vector, which makes your data much more readable: ```{r} # Give each value a name corresponding to the opponent opponents <- c("BAL", "CIN", "ATL", "LAC", "NO", "LV", "TB", "DEN", "CAR", "LV") names(mahomes_yards) <- opponents mahomes_yards ``` ```{r} # Now you can see which game each number belongs to # Access by name mahomes_yards["BAL"] mahomes_yards["TB"] ``` ::: callout-tip ## Why Name Your Vectors? Names make vectors much easier to read and interpret. When you come back to your code weeks later, `mahomes_yards["BAL"]` is much clearer than `mahomes_yards[1]`. ::: ------------------------------------------------------------------------ ## Part 2: Missing Values — `NA` `NA` stands for **Not Available**. It is R's placeholder for missing data. Think of it like a blank cell in a spreadsheet — the data should be there, but it isn't. ```{r} # Lamar Jackson's rushing yards per game, 2024 season # He missed 2 games due to injury (games 6 and 11) jackson_rushing <- c(45, 62, 38, 71, 55, NA, 49, 83, 60, 41, NA, 77) jackson_rushing ``` ```{r} # R tells you there are 2 NAs summary(jackson_rushing) ``` ### Detecting Missing Values with `is.na()` ```{r} # Which positions are NA? is.na(jackson_rushing) ``` ```{r} # How many games did he miss? sum(is.na(jackson_rushing)) ``` ```{r} # How many games did he play? sum(!is.na(jackson_rushing)) # ! means "NOT" ``` ::: callout-warning ## ️ NAs are Contagious! If any value in your calculation is `NA`, the result is also `NA`: ```{r} mean(jackson_rushing) # Returns NA — because of the missing games! ``` This seems annoying, but it's actually protecting you. R is saying: "I don't know the answer because some data is missing." ::: ------------------------------------------------------------------------ ## Part 3: Working Around Missing Values ### `na.rm = TRUE` Most R statistical functions have an `na.rm` argument. Setting it to `TRUE` tells R to **remove the NAs before calculating**. ```{r} # Now we get a real answer mean(jackson_rushing, na.rm = TRUE) ``` ```{r} sum(jackson_rushing, na.rm = TRUE) max(jackson_rushing, na.rm = TRUE) min(jackson_rushing, na.rm = TRUE) ``` ::: callout-note ## What does `na.rm` stand for? `na.rm` = **NA remove**. Setting it to `TRUE` removes `NA`s before doing the calculation. ::: ### Removing NAs Entirely Sometimes you want a clean vector with no `NA`s at all: ```{r} # na.omit() removes all NA values jackson_clean <- na.omit(jackson_rushing) jackson_clean length(jackson_clean) # 10 games (the 2 missed are gone) ``` ```{r} # Alternatively, use logical indexing (we'll cover this more in Lab 6) jackson_played <- jackson_rushing[!is.na(jackson_rushing)] jackson_played ``` ### Replacing NAs with a Value Sometimes you want to replace `NA` with something meaningful (like 0 for a missed game): ```{r} jackson_with_zeros <- jackson_rushing jackson_with_zeros[is.na(jackson_with_zeros)] <- 0 jackson_with_zeros ``` ::: callout-important ## Should You Replace NAs with Zero? It depends on the context! For rushing yards, replacing with 0 makes sense (if he didn't play, he gained 0 yards). But for body temperature or a test score, replacing missing data with 0 would be **wrong and misleading**. In science, always document how you handled missing values and justify your choice. ::: ------------------------------------------------------------------------ ## Part 4: Vector Arithmetic Like we saw with sequences, arithmetic on vectors applies to **every element**: ```{r} # Points scored by the Dallas Cowboys in each week of the season cowboys_pts <- c(24, 17, 31, 20, 28, 14, 35, 22, 19, 27, 30, 21, 16, 28, 33, 18, 24) # Points scored by opponents (what they allowed) cowboys_allowed <- c(20, 21, 28, 17, 31, 20, 28, 30, 22, 14, 17, 24, 27, 21, 30, 27, 20) # Point differential for each game (positive = win, negative = loss) point_diff <- cowboys_pts - cowboys_allowed point_diff ``` ```{r} # Which games did they win? won_game <- point_diff > 0 won_game ``` ```{r} # How many wins? sum(won_game) ``` ```{r} # How many losses? sum(point_diff < 0) ``` ```{r} # Total points scored this season sum(cowboys_pts) # Average points per game round(mean(cowboys_pts), 1) ``` ### Arithmetic Between Two Vectors ```{r} # Suppose each point is worth different fantasy values # 1 point = 1 fantasy point, but bonus points for games over 28 bonus <- ifelse(cowboys_pts > 28, 3, 0) # 3 bonus points for high-scoring games fantasy_score <- cowboys_pts + bonus fantasy_score ``` ::: callout-tip ## `ifelse()` — Your First Conditional Function `ifelse(test, value_if_TRUE, value_if_FALSE)` applies a test to every element and returns one of two values. This is extremely useful and you'll use it constantly when cleaning data. ::: ------------------------------------------------------------------------ ## Part 5: Putting It Together — QB Comparison Let's compare two quarterbacks across a shared 10-game stretch, handling missing data properly: ```{r} # 10-game passer rating comparison qb_names <- c("Game 1", "Game 2", "Game 3", "Game 4", "Game 5", "Game 6", "Game 7", "Game 8", "Game 9", "Game 10") # Patrick Mahomes passer ratings (0-158.3 scale) mahomes_rating <- c(112.4, 98.7, 131.2, NA, 105.8, 122.1, 89.3, 141.5, 107.6, 118.9) # Jalen Hurts passer ratings hurts_rating <- c(98.1, 107.5, 88.4, 125.3, 110.2, NA, 119.7, 92.8, NA, 130.1) # How many games did each miss? cat("Mahomes missed:", sum(is.na(mahomes_rating)), "game(s)\n") cat("Hurts missed: ", sum(is.na(hurts_rating)), "game(s)\n") ``` ```{r} # Average passer rating (ignoring missed games) mahomes_avg <- mean(mahomes_rating, na.rm = TRUE) hurts_avg <- mean(hurts_rating, na.rm = TRUE) cat("Mahomes average passer rating:", round(mahomes_avg, 1), "\n") cat("Hurts average passer rating: ", round(hurts_avg, 1), "\n") cat("Difference:", round(mahomes_avg - hurts_avg, 1), "points\n") ``` ```{r} # Simple side-by-side bar chart barplot( c(mahomes_avg, hurts_avg), names.arg = c("P. Mahomes", "J. Hurts"), col = c("#E31837", "#004C54"), # Chiefs red, Eagles midnight green ylim = c(0, 160), ylab = "Average Passer Rating", main = "QB Passer Rating Comparison\n(missing games excluded)", border = "white" ) abline(h = 100, col = "gray50", lty = 2) text(0.7, 102, "100 = Good QB Rating", col = "gray30", cex = 0.8) ``` ------------------------------------------------------------------------ ## 3-Minute Knowledge Check *Close your notes. Answer these on your own — you have 3 minutes. We'll go through the answers together.* ::: callout-caution ## Knowledge Check Questions **1.** What does `NA` stand for in R? a) Not Applicable    b) Not Available    c) Null Answer    d) Negative Amount **2.** You run `mean(c(10, 20, NA, 40))` and get `NA`. How do you fix this? **3.** What does `sum(is.na(x))` calculate? a) The sum of all values in x    b) The number of missing values in x    c) Whether x has any NAs    d) The mean of x **4.** True or False: `na.omit(x)` replaces NAs with zero. **5.** You have a vector `scores <- c(88, NA, 91, 75, NA)`. Write the code to calculate the mean ignoring the missing values. ::: ::: {.callout-caution collapse="true"} ## Answers (Only reveal if you have answered the above questions) **1.** b) Not Available — `NA` is R's way of representing missing or unknown data. **2.** Add `na.rm = TRUE`: `mean(c(10, 20, NA, 40), na.rm = TRUE)` → returns `23.33`. The `na.rm` argument tells R to **remove NAs** before calculating. **3.** b) The number of missing values in x — `is.na(x)` returns a logical vector of TRUE/FALSE, and `sum()` counts the TRUEs (each TRUE = 1). **4.** **False** — `na.omit(x)` **removes** the NA values entirely, returning a shorter vector. To replace NAs with zero, you would use `x[is.na(x)] <- 0`. **5.** `mean(scores, na.rm = TRUE)` → returns `84.67` (average of 88, 91, and 75). ::: ------------------------------------------------------------------------ ## Lab 4 Checklist Before you leave, make sure you can: - [ ] Create a named vector using `c()` and `names()` - [ ] Explain what `NA` means in R - [ ] Use `is.na()` to find missing values - [ ] Use `sum(is.na())` to count missing values - [ ] Calculate `mean()`, `sum()`, etc. with `na.rm = TRUE` - [ ] Use `na.omit()` to remove NAs - [ ] Replace NAs using logical indexing `x[is.na(x)] <- value` - [ ] Perform arithmetic between two vectors ::: callout-tip ## Bonus Challenge Find the season statistics for two NFL quarterbacks (any two you like). Create named vectors for their touchdown passes in each week. One of them should have at least one `NA` for a missed game. Compare their averages using `na.rm = TRUE` and make a barplot. Add this to your Quarto document and render it. ::: ------------------------------------------------------------------------ ## Before Next Class (Friday) - Read **Intro2r §3.1, 3.2** - Friday (Lab 5) we move to **data frames and matrices** — where rows and columns begin - Practice: create a vector with at least one `NA` and calculate its mean both with and without `na.rm = TRUE` ------------------------------------------------------------------------