Day 10 Activity: Summarizing the penguins

Monday, June 1, 2026

Author

Your Name Here

Setup

library(tidyverse)
library(palmerpenguins)

Today’s activity uses the penguins dataset (palmerpenguins package). Glimpse it before you start:

glimpse(penguins)
Rows: 344
Columns: 8
$ species           <fct> Adelie, Adelie, Adelie, Adelie, Adelie, Adelie, Adel…
$ island            <fct> Torgersen, Torgersen, Torgersen, Torgersen, Torgerse…
$ bill_length_mm    <dbl> 39.1, 39.5, 40.3, NA, 36.7, 39.3, 38.9, 39.2, 34.1, …
$ bill_depth_mm     <dbl> 18.7, 17.4, 18.0, NA, 19.3, 20.6, 17.8, 19.6, 18.1, …
$ flipper_length_mm <int> 181, 186, 195, NA, 193, 190, 181, 195, 193, 190, 186…
$ body_mass_g       <int> 3750, 3800, 3250, NA, 3450, 3650, 3625, 4675, 3475, …
$ sex               <fct> male, female, female, NA, female, male, female, male…
$ year              <int> 2007, 2007, 2007, 2007, 2007, 2007, 2007, 2007, 2007…

We will practice using two data verbs, summarize() and group_by(), as well as reduction functions like n() and mean().

Part 1

For each operation below, say which family it belongs to: reduction (variable to one number), transformation (variable to one value per row), or neither (a whole-table data verb).

Fill in the third column.
# Operation Family
1 The largest body mass in the dataset [Your answer.]
2 Body mass converted from grams to kilograms [Your answer.]
3 The number of penguins [Your answer.]
4 Whether each penguin weighs more than 4000 g [Your answer.]
5 The number of distinct islands [Your answer.]
6 Labeling each penguin "heavy" or "light" [Your answer.]
7 Keeping only the Gentoo penguins [Your answer.]
8 The standard deviation of flipper length [Your answer.]
Tip

The fastest test: does it return one number (reduction), one per row (transformation), or a whole new table (neither, that’s a data verb)? Number 7 is tricky.

Part 2

2a. Write a single summarize() that returns the number of penguins, the mean body mass, and the mean flipper length.

# Your code:

2b. Run this, then explain in one sentence why the result is what it is:

penguins |>
  summarize(avg_mass = mean(body_mass_g))
# A tibble: 1 × 1
  avg_mass
     <dbl>
1       NA

[Your explanation here.]

2c. Fix 2b so it returns a real number. What one change did you make, and why does it work?

# Your fixed code:

[Your explanation here.]

Part 3

For 3a–3d, first write down your prediction for how many rows the result will have and which columns it will contain. Then run it and check.

3a.

penguins |>
  group_by(species) |>
  summarize(n = n(), avg_mass = mean(body_mass_g, na.rm = TRUE))

Prediction (rows / columns): [Your answer.]

Actual: [Your answer.]

3b.

penguins |>
  group_by(species, sex) |>
  summarize(n = n(), avg_mass = mean(body_mass_g, na.rm = TRUE),
            .groups = "drop")

Prediction (rows / columns): [Your answer.]

Actual: [Your answer.]

Did anything about the number of rows surprise you? (Look closely at the sex column.)

[Your answer.]

3c. This one is a trap. Predict before you run it, then run it:

penguins |>
  group_by(body_mass_g) |>
  summarize(n = n(), .groups = "drop")

Prediction (rows / columns): [Your answer.]

Actual: [Your answer.]

What went wrong, and what does it tell you about which kinds of variables you should group by?

[Your answer.]

3d. Without grouping at all:

penguins |>
  summarize(n = n())

Prediction (rows / columns): [Your answer.]

Actual: [Your answer.]

Explain the relationship between 3a, 3b, and 3d in terms of “what does one row of the output represent?”

[Your answer.]

Part 4 — Don’t trust a mean without its n()

Run the finest grouping we can make from these three variables:

penguins |>
  group_by(species, island, sex) |>
  summarize(n = n(), avg_mass = mean(body_mass_g, na.rm = TRUE),
            .groups = "drop")

4a. Which group has the fewest penguins? Would you be comfortable quoting that group’s avg_mass in a report? Why or why not?

[Your answer.]

4b. In one or two sentences: why is n = n() worth adding to every group summary you write, even when nobody asked for a count?

[Your answer.]

Code appendix

library(tidyverse)
library(palmerpenguins)
glimpse(penguins)
# Your code:

penguins |>
  summarize(avg_mass = mean(body_mass_g))
# Your fixed code:

penguins |>
  group_by(species) |>
  summarize(n = n(), avg_mass = mean(body_mass_g, na.rm = TRUE))
penguins |>
  group_by(species, sex) |>
  summarize(n = n(), avg_mass = mean(body_mass_g, na.rm = TRUE),
            .groups = "drop")
penguins |>
  group_by(body_mass_g) |>
  summarize(n = n(), .groups = "drop")
penguins |>
  summarize(n = n())
penguins |>
  group_by(species, island, sex) |>
  summarize(n = n(), avg_mass = mean(body_mass_g, na.rm = TRUE),
            .groups = "drop")