Chart Types

🎨 Base R Visualization: Chart Types

The Art Gallery Analogy

Imagine you’re a museum curator. You have amazing stories to tell, but you need to pick the right frame for each painting. Some stories need a line to show a journey. Others need bars to compare heroes. Some need dots scattered like stars to find hidden patterns.

R’s plot functions are your frames. Each chart type tells your data’s story in a different way. Let’s explore your gallery!

🖼️ Plot Function Basics

The Magic Paintbrush: plot()

Think of plot() like a magic crayon that can draw almost anything. You give it numbers, and it creates pictures!

The simplest drawing:

# Just give it some numbers
x <- c(1, 2, 3, 4, 5)
y <- c(2, 4, 6, 8, 10)
plot(x, y)

What happens? R looks at your numbers and picks the best picture type automatically!

The Three Magic Words

Every plot understands these basic commands:

Example:

plot(x, y,
     main = "My First Chart",
     xlab = "Days",
     ylab = "Cookies Eaten")

Quick Tips 🚀

• No data? plot() will show an error
• One number list? It uses position as x
• Two lists? First is x, second is y

📈 Line Plots

The Story of a Journey

A line plot is like connecting dots on a treasure map. It shows how something changes over time or moves from one point to another.

When to use it?

• Temperature changes during the day
• Your height as you grow up
• Stock prices over months

Creating Line Plots

Method 1: Using type = "l"

days <- 1:7
temperature <- c(20, 22, 19, 25, 28, 26, 24)

plot(days, temperature,
     type = "l",
     main = "Week's Temperature",
     xlab = "Day",
     ylab = "Temp (°C)")

Method 2: The lines() function

# First create an empty plot
plot(days, temperature, type = "n")
# Then add the line
lines(days, temperature, col = "blue")

Line Styles You Can Use

Try this:

plot(1:5, c(1,3,2,5,4),
     type = "b",
     col = "red",
     lwd = 2)  # lwd = line width

🔵 Scatter Plots

The Star Map

Imagine throwing a handful of stars onto black paper. Where each star lands tells you something. Scatter plots show relationships between two things - like friends standing together based on how much they like pizza AND ice cream!

When to use it?

• Height vs. Weight
• Study hours vs. Test scores
• Age vs. Shoe size

Creating Scatter Plots

height <- c(150, 160, 155, 170, 165)
weight <- c(50, 65, 55, 75, 60)

plot(height, weight,
     main = "Height vs Weight",
     xlab = "Height (cm)",
     ylab = "Weight (kg)",
     pch = 19,  # Solid circles
     col = "blue")

Point Shapes (pch)

The pch argument changes your dot shapes:

Pro tip: Use cex to change point size:

plot(x, y, pch = 19, cex = 2)  # 2x bigger

Reading Scatter Patterns

graph TD
    A["Points going up-right?"] --> B["Positive relationship!"]
    C["Points going down-right?"] --> D["Negative relationship!"]
    E["Points everywhere?"] --> F["No relationship"]

📊 Bar Plots

The Podium of Champions

Bar plots are like the Olympic podium - they compare things side by side! Each bar’s height shows how big or important something is.

When to use it?

• Comparing scores
• Counting categories (apples vs oranges)
• Showing survey results

Creating Bar Plots

fruits <- c(5, 8, 3, 10)
names(fruits) <- c("Apple", "Banana",
                   "Cherry", "Date")

barplot(fruits,
        main = "Fruit Count",
        xlab = "Fruit",
        ylab = "Count",
        col = "skyblue")

Horizontal Bars

Sometimes it’s easier to read sideways:

barplot(fruits,
        horiz = TRUE,
        main = "Fruit Count",
        col = rainbow(4))

Side-by-Side Bars

Compare groups with matrices:

data <- matrix(c(5,3,8,6,4,7),
               nrow = 2)
colnames(data) <- c("Mon","Tue","Wed")
rownames(data) <- c("Boys","Girls")

barplot(data,
        beside = TRUE,
        col = c("blue","pink"),
        legend = rownames(data))

📉 Histograms

The Sorting Hat

A histogram is like sorting students into houses based on their height. It counts how many fall into each “bin” (range).

Key difference from bar plots:

• Bar plots: Compare categories (apples, bananas)
• Histograms: Show distribution of numbers (how many people are 150-160cm tall?)

Creating Histograms

ages <- c(12,15,13,18,14,16,15,17,
          14,13,16,15,14,17,16)

hist(ages,
     main = "Age Distribution",
     xlab = "Age",
     ylab = "Frequency",
     col = "lightgreen",
     border = "darkgreen")

Controlling Bins

# More bins = more detail
hist(ages, breaks = 10)

# Specific breakpoints
hist(ages, breaks = c(12,14,16,18))

What Histograms Tell You

graph TD
    A["Shape tells the story"]
    A --> B["Bell shape = Normal"]
    A --> C["Leaning left = Right-skewed"]
    A --> D["Leaning right = Left-skewed"]
    A --> E["Two humps = Bimodal"]

📦 Box Plots

The Five-Number Summary Portrait

A box plot is like a magic portrait that shows 5 important facts about your data in one picture:

1. Minimum - The smallest (excluding outliers)
2. Q1 - 25% of data is below this
3. Median - The middle value
4. Q3 - 75% of data is below this
5. Maximum - The largest (excluding outliers)

Creating Box Plots

scores <- c(65, 70, 75, 80, 85,
            90, 95, 40, 100)

boxplot(scores,
        main = "Test Scores",
        ylab = "Score",
        col = "gold")

Comparing Groups

# Multiple box plots side by side
class_A <- c(75, 80, 82, 78, 85)
class_B <- c(60, 65, 70, 68, 72)
class_C <- c(90, 92, 88, 95, 91)

boxplot(class_A, class_B, class_C,
        names = c("A", "B", "C"),
        main = "Class Comparison",
        col = c("red","green","blue"))

Reading Box Plots

📐 QQ Plots

The Normal Detector

Q-Q stands for Quantile-Quantile. Think of it as a detective tool that checks if your data follows the “normal” bell curve pattern.

Why does this matter? Many statistics only work correctly if your data is “normal” (follows a bell curve). QQ plots help you check!

Creating QQ Plots

# Generate some data
my_data <- rnorm(100)  # Normal data

# Create QQ plot
qqnorm(my_data,
       main = "Normal QQ Plot")
qqline(my_data, col = "red")

Reading QQ Plots

The Rule: If points follow the red line = Normal! ✅

graph TD
    A["Look at the points"]
    A --> B["Follow the line?"]
    B --> C["YES = Normal data ✓"]
    B --> D["NO = Not normal ✗"]
    D --> E["Curved at ends = Skewed"]
    D --> F["S-shaped = Heavy tails"]

Example: Normal vs Non-Normal

# Normal data - points on line
normal <- rnorm(100)
qqnorm(normal, main = "Normal")
qqline(normal)

# Skewed data - points curve away
skewed <- rexp(100)  # Exponential
qqnorm(skewed, main = "Skewed")
qqline(skewed)

🎯 Quick Reference: Which Chart When?

🌟 Your First Gallery

You now have six powerful frames for your data stories:

1. Line plots - The journey teller
2. Scatter plots - The relationship finder
3. Bar plots - The champion comparer
4. Histograms - The sorting hat
5. Box plots - The five-fact portrait
6. QQ plots - The normal detector

Each one has a special purpose. Choose wisely, and your data will speak clearly to everyone who looks at your charts!

Remember: The best chart is the one that tells your story simply and honestly. 🎨