---
title: "Lab 6: Subsetting Vectors, Matrices & Data Frames"
subtitle: "BI 255 · Bethune-Cookman University · Fall 2026"
author: "Dr. Rosie Stanbrook-Buyer"
date: "August 31, 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)
```
------------------------------------------------------------------------
## Filtering NBA Data Like a Pro
Imagine you're an NBA analyst and you have stats on every player in the league. You don't need all 450+ players at once — you need just the All-Stars, or just the players who average over 20 points, or just the centers on Western Conference teams.
**Subsetting** is how you extract exactly the slice of data you need. It is one of the most important skills in data science.
::: callout-note
## Learning Objectives
By the end of this lab you will be able to:
1. Subset vectors using numeric indices `[ ]`
2. Subset using logical conditions
3. Use `which()` to find positions matching a condition
4. Subset data frames by row, column, and condition
5. Use `subset()` for readable filtering
6. Understand negative indexing (removing elements)
:::
------------------------------------------------------------------------
## Part 1: Subsetting Vectors
### By Position (Numeric Index)
```{r}
# Top 10 NBA scorers' points per game (2025-26 season, simulated)
top_scorers_ppg <- c(32.1, 29.8, 27.4, 26.9, 25.3,
24.8, 23.7, 22.9, 21.4, 20.8)
top_scorers_names <- c("Luka Doncic", "Jayson Tatum", "Giannis Antetokounmpo",
"Stephen Curry", "LeBron James",
"Kevin Durant", "Devin Booker", "Donovan Mitchell",
"Ja Morant", "Joel Embiid")
names(top_scorers_ppg) <- top_scorers_names
top_scorers_ppg
```
```{r}
# Who is the #1 scorer?
top_scorers_ppg[1]
```
```{r}
# Top 3 scorers
top_scorers_ppg[1:3]
```
```{r}
# 5th, 7th, and 9th scorers
top_scorers_ppg[c(5, 7, 9)]
```
### Negative Indexing — Removing Elements
```{r}
# Everyone EXCEPT the first player
top_scorers_ppg[-1]
```
```{r}
# Remove the top 3 (show everyone else)
top_scorers_ppg[-c(1, 2, 3)]
```
::: callout-tip
## Negative Indexing
Put a minus sign before an index to **exclude** that position. This is a quick way to drop specific elements without having to retype everything else.
:::
------------------------------------------------------------------------
## Part 2: Subsetting with Logical Conditions
This is where subsetting gets really powerful. You can extract elements based on whether they meet a condition.
```{r}
# Which players average more than 25 PPG?
top_scorers_ppg[top_scorers_ppg > 25]
```
```{r}
# Which players average between 22 and 26 PPG?
top_scorers_ppg[top_scorers_ppg >= 22 & top_scorers_ppg <= 26]
```
::: callout-note
## How Logical Subsetting Works
`top_scorers_ppg > 25` creates a logical vector:
```
TRUE TRUE TRUE TRUE FALSE FALSE FALSE FALSE FALSE FALSE
```
When you put that inside `[ ]`, R returns only the elements where the value is `TRUE`.
This is one of the most important patterns in R — **you'll use it in almost every analysis.**
:::
```{r}
# Using which() to get the positions of matching elements
which(top_scorers_ppg > 25)
```
```{r}
# Useful for getting names
names(which(top_scorers_ppg > 25))
```
------------------------------------------------------------------------
## Part 3: Subsetting Data Frames
Let's work with a fuller NBA dataset:
```{r}
# NBA player stats for a subset of the 2025-26 season
nba <- data.frame(
player = c("Luka Doncic", "Jayson Tatum", "Giannis Antetokounmpo",
"Stephen Curry", "LeBron James", "Kevin Durant",
"Devin Booker", "Donovan Mitchell", "Ja Morant",
"Joel Embiid", "Nikola Jokic", "Shai Gilgeous-Alexander",
"Tyrese Haliburton", "Anthony Edwards", "Bam Adebayo"),
team = c("DAL", "BOS", "MIL", "GSW", "LAL", "PHX",
"PHX", "CLE", "MEM", "PHI", "DEN", "OKC",
"IND", "MIN", "MIA"),
conference = c("W", "E", "E", "W", "W", "W",
"W", "E", "W", "E", "W", "W",
"E", "W", "E"),
position = c("PG", "SF", "PF", "PG", "SF", "SF",
"SG", "SG", "PG", "C", "C", "SG",
"PG", "SG", "C"),
ppg = c(32.1, 29.8, 27.4, 26.9, 25.3, 24.8,
23.7, 22.9, 21.4, 20.8, 26.4, 30.1,
22.7, 25.9, 19.8),
rpg = c(8.7, 8.1, 11.5, 4.3, 7.2, 6.8,
4.2, 4.5, 5.9, 11.8, 12.4, 4.6,
3.9, 5.3, 10.7),
apg = c(9.1, 4.8, 5.7, 6.4, 8.3, 4.1,
4.9, 4.2, 8.1, 3.4, 9.2, 6.3,
10.8, 5.6, 3.5),
is_allstar = c(TRUE, TRUE, TRUE, TRUE, TRUE, TRUE,
FALSE, TRUE, FALSE, TRUE, TRUE, TRUE,
FALSE, TRUE, TRUE)
)
```
### Subset by Row Index
```{r}
# First 3 players
nba[1:3, ]
```
### Subset by Column
```{r}
# Just names and PPG
nba[, c("player", "ppg")]
```
### Subset by Condition — Rows Matching a Criterion
```{r}
# All Western Conference players
nba[nba$conference == "W", ]
```
```{r}
# All-Stars only
nba[nba$is_allstar == TRUE, c("player", "team", "ppg")]
```
```{r}
# Players averaging over 25 PPG in the Eastern Conference
nba[nba$ppg > 25 & nba$conference == "E", ]
```
### Using `subset()` — More Readable
```{r}
# subset() lets you skip the $ and write conditions more cleanly
subset(nba, ppg > 25, select = c(player, team, ppg, conference))
```
```{r}
# Centers (position == "C") with more than 10 rebounds per game
subset(nba, position == "C" & rpg > 10)
```
::: callout-tip
## `[ ]` vs `subset()` — Which Should You Use?
Both work, but:
- `subset()` is more **readable** — great when you're writing code others (me!) will see
- `[ ]` is more **flexible** — required for some advanced operations
- In the next unit, we'll learn `dplyr::filter()` which is even cleaner!
For now, practice both. They produce the same result.
:::
------------------------------------------------------------------------
## Part 4: Modifying Subsets
You can also use subsetting to **change** specific values:
```{r}
# Suppose Joel Embiid's PPG needs to be corrected to 21.2
nba$ppg[nba$player == "Joel Embiid"] <- 21.2
# Verify the change
nba[nba$player == "Joel Embiid", c("player", "ppg")]
```
```{r}
# Update all Western Conference player records to add a "W" flag
nba$west_flag <- nba$conference == "W"
head(nba[, c("player", "conference", "west_flag")])
```
------------------------------------------------------------------------
## Part 5: Subsetting Matrices
```{r}
# Create a stats matrix (PPG, RPG, APG for 5 players)
stat_matrix <- matrix(
c(32.1, 8.7, 9.1,
29.8, 8.1, 4.8,
27.4, 11.5, 5.7,
26.9, 4.3, 6.4,
30.1, 4.6, 6.3),
nrow = 5,
byrow = TRUE
)
rownames(stat_matrix) <- c("Doncic", "Tatum", "Giannis", "Curry", "SGA")
colnames(stat_matrix) <- c("PPG", "RPG", "APG")
stat_matrix
```
```{r}
# Row subsetting
stat_matrix["Curry", ]
```
```{r}
# Column subsetting
stat_matrix[, "PPG"]
```
```{r}
# Who has the highest PPG?
rownames(stat_matrix)[which.max(stat_matrix[, "PPG"])]
```
------------------------------------------------------------------------
## Part 6: Quick Summary Analysis
Let's use subsetting to answer real questions about our NBA dataset:
```{r}
# Question 1: How many All-Stars are there?
sum(nba$is_allstar)
```
```{r}
# Question 2: What is the average PPG for All-Stars vs non-All-Stars?
allstar_ppg <- mean(nba$ppg[nba$is_allstar == TRUE])
non_allstar_ppg <- mean(nba$ppg[nba$is_allstar == FALSE])
cat("All-Star average PPG: ", round(allstar_ppg, 1), "\n")
cat("Non-All-Star average PPG:", round(non_allstar_ppg, 1), "\n")
```
```{r}
# Question 3: Which conference has higher average PPG?
east_ppg <- mean(nba$ppg[nba$conference == "E"])
west_ppg <- mean(nba$ppg[nba$conference == "W"])
cat("Eastern Conference avg PPG:", round(east_ppg, 1), "\n")
cat("Western Conference avg PPG:", round(west_ppg, 1), "\n")
```
```{r}
# Simple visualization: PPG by conference
boxplot(ppg ~ conference, data = nba,
col = c("#007AC1", "#C8102E"), # NBA blue and red
xlab = "Conference",
ylab = "Points Per Game",
main = "PPG Distribution by NBA Conference\n(Selected Players, 2025–26)")
```
------------------------------------------------------------------------
## 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.** You have a vector `x <- c(10, 20, 30, 40, 50)`. What does `x[3]` return?
a) `10` b) `20` c) `30` d) `3`
**2.** What does `x[-2]` return from the same vector?
a) `20` b) `10 20 30 40 50` c) `10 30 40 50` d) An error
**3.** You have a data frame `df` with a column `score`. Write the code to keep only rows where `score > 90`.
**4.** What does `which(x > 25)` return for `x <- c(10, 20, 30, 40, 50)`?
a) `30 40 50` b) `3 4 5` c) `TRUE FALSE TRUE TRUE TRUE` d) `FALSE FALSE TRUE TRUE TRUE`
**5.** True or False: `subset(nba, ppg > 25)` and `nba[nba$ppg > 25, ]` produce the same result.
:::
::: {.callout-caution collapse="true"}
## Answers (Only reveal if you have answered the above questions)
**1.** c) `30` — R uses 1-based indexing, so `x[1]` = 10, `x[2]` = 20, `x[3]` = 30.
**2.** c) `10 30 40 50` — The minus sign **excludes** that position. `x[-2]` means "everything except the 2nd element," which removes 20.
**3.** `df[df$score > 90, ]` — Note the comma and the empty column position (meaning "all columns"). You could also write `subset(df, score > 90)`.
**4.** b) `3 4 5` — `which()` returns the **positions** (indices) where the condition is TRUE, not the values themselves. Positions 3, 4, and 5 contain values 30, 40, and 50.
**5.** **True** — Both approaches filter rows to those where `ppg > 25`. `subset()` is just more readable. In Lab 9 we'll learn `filter()` from dplyr, which is even cleaner.
:::
------------------------------------------------------------------------
## Lab 6 Checklist
Before you leave, make sure you can:
- [ ] Select specific elements from a vector by position
- [ ] Use negative indexing to exclude elements
- [ ] Subset using a logical condition (e.g., `x[x > 10]`)
- [ ] Use `which()` to find positions matching a condition
- [ ] Subset a data frame by row and/or column
- [ ] Filter a data frame using a condition on a column (`df[df$col > value, ]`)
- [ ] Use `subset()` for readable filtering
- [ ] Modify values in a subset
::: callout-tip
## Bonus Challenge
Using the `nba` data frame, find all players whose assists per game (`apg`) is higher than their rebounds per game (`rpg`). Print their names, team, and both stats. Which position do most of these players play?
:::
------------------------------------------------------------------------
## Before Next Class (Wednesday)
- Read **Intro2r Ch. 3 §3.2–3.3**
- Wednesday (Lab 7): Dates & Times, plus the apply family of functions
- Remember: **Quiz 2 is this Friday (Sep 4)**
::: callout-important
## Quiz 2 Reminder
Quiz 2 covers **Labs 3, 4, 5, and 6**:
- Sequences (`seq()`, `rep()`, `:`)
- Vectors and missing values (`NA`, `is.na()`, `na.rm`)
- Data frames (creating, exploring, accessing)
- Subsetting (`[ ]`, `subset()`, logical conditions)
:::
------------------------------------------------------------------------
