Week 1: What is Data Science?

CDAE 7990 · Applied Data Science & Visualization

Andrew Van Leuven

2026-09-03

DRAFT

This is a placeholder slide deck. Content is not finalized.

How these slides work

• Each ## heading creates a new slide
• Use # Heading (single #) for a section title slide
• Add {background-color="#154734"} to a section slide for a full-color break
• Speaker notes go in a ::: {.notes} block (visible in presenter view, not to audience)

This is a speaker note. Press S in the browser to open presenter view.

Incremental bullet points

Use . . . between items to reveal them one at a time:

• First point

• Second point (appears on next click)

• Third point (appears on next click)

Two-column layout

Left column

Use this for text alongside a figure, or to compare two things side by side.

Right column

# code can go here too
library(tidyverse)

dplyr basics

The core verbs for data manipulation

The pipe operator

The pipe |> passes the left-hand object into the first argument of the right-hand function.

# Without pipe
filter(mpg, cyl == 4)

# With pipe — reads like a sentence
mpg |> filter(cyl == 4)

Chain multiple steps together:

mpg |>
  filter(cyl == 4) |>
  select(manufacturer, model, hwy) |>
  arrange(desc(hwy))

filter()

Keep rows that match a condition.

library(tidyverse)

# Single condition
mpg |> filter(year == 2008)

# Multiple conditions (AND)
mpg |> filter(year == 2008, cyl == 4)

# OR condition
mpg |> filter(class == "suv" | class == "pickup")

# Useful shorthand for OR on one variable
mpg |> filter(class %in% c("suv", "pickup"))

select()

Keep (or drop) columns by name.

# Keep specific columns
mpg |> select(manufacturer, model, year, hwy)

# Drop a column with -
mpg |> select(-model)

# Select a range
mpg |> select(manufacturer:year)

# Rename while selecting
mpg |> select(make = manufacturer, highway_mpg = hwy)

mutate()

Create new columns or overwrite existing ones.

mpg |>
  mutate(
    hwy_kpl = hwy * 0.425,           # highway km per liter
    efficient = hwy > 30             # logical column
  ) |>
  select(manufacturer, model, hwy, hwy_kpl, efficient)

summarize() + group_by()

Collapse rows to summary statistics, optionally by group.

# Overall mean
mpg |> summarize(mean_hwy = mean(hwy))

# By group
mpg |>
  group_by(class) |>
  summarize(
    mean_hwy = mean(hwy),
    n = n()
  ) |>
  arrange(desc(mean_hwy))

ggplot2 basics

The grammar of graphics

The ggplot2 template

Every plot follows the same structure:

ggplot(data = <DATA>, aes(x = <X>, y = <Y>)) +
  geom_<TYPE>()

• ggplot() initializes the plot and sets the data
• aes() maps variables to visual properties (aesthetics)
• geom_*() determines the type of plot

Scatter plot

ggplot(data = mpg, aes(x = displ, y = hwy)) +
  geom_point()

Add color by a third variable:

ggplot(data = mpg, aes(x = displ, y = hwy, color = class)) +
  geom_point()

Bar chart

# Count of cars by class
ggplot(data = mpg, aes(x = class)) +
  geom_bar()

# Pre-summarized data: use geom_col()
mpg |>
  group_by(class) |>
  summarize(mean_hwy = mean(hwy)) |>
  ggplot(aes(x = class, y = mean_hwy)) +
  geom_col()

Labels + theme

ggplot(data = mpg, aes(x = displ, y = hwy, color = class)) +
  geom_point() +
  labs(
    title = "Fuel efficiency vs. engine size",
    x = "Engine displacement (liters)",
    y = "Highway MPG",
    color = "Vehicle class"
  ) +
  theme_minimal()

Facets

Split one plot into small multiples by a variable.

ggplot(data = mpg, aes(x = displ, y = hwy)) +
  geom_point() +
  facet_wrap(~ class)

Two-variable facet grid:

ggplot(data = mpg, aes(x = displ, y = hwy)) +
  geom_point() +
  facet_grid(drv ~ cyl)