---
title: "Lab 2: a structured reproducible report"
subtitle: "Reproducible Workflows · Session 1"
author: "Your Name"
format: html
editor: source
---

> ~30 min. Build a short **report** that follows the 5-section structure from the
> slides. The point is **structure and captions**, not length; a clean report
> beats a long one. We use the penguins data from the `{palmerpenguins}`
> package, so this renders anywhere with that package installed.

## The question

Write **one or two sentences** stating the question this report answers. For
example: *"Do penguin species differ in flipper length, and by how much?"*

Put it under a `## The question` heading.

## The data

Under a `## The data` heading, add a short paragraph and a chunk that *loads*
the data (this is the transparent part; the reader sees exactly what you used):

```{r}
#| label: data
#| message: false
library(palmerpenguins)   # 344 penguins, 3 species, body & flipper measurements
library(tidyverse)        # ggplot2 + dplyr for the figure and table below
head(penguins)
```

## Cleaning

Even clean data gets a note. The `penguins` data arrives tidy and typed, but a
few rows have missing measurements. Under `## Cleaning`, write what *you*
decided and why, e.g. *"no cleaning beyond the built-in missing values; the
figure drops them and I remove them when summarising."* Deciding deliberately
and writing it down is the reproducible habit; you'll do much more of this in
Session 2.

## Results

Under `## Results`:

1. A **figure** with a real caption. Build it with `ggplot2`:

```{r}
#| label: fig-flipper
#| fig-cap: "Distribution of flipper length by penguin species."
ggplot(penguins, aes(species, flipper_length_mm)) +
  geom_boxplot(fill = c("#2E7D6B", "#E8A33D", "#4A5899")) +
  ylab("Flipper length (mm)")
```

2. A **table** that summarises the numbers behind the figure, also captioned
(note `na.rm = TRUE`, matching your cleaning note):

```{r}
#| label: tbl-flipper-summary
#| tbl-cap: "Flipper length (mm) by species: mean and standard deviation."
penguins |>
  group_by(species) |>
  summarise(
    mean = mean(flipper_length_mm, na.rm = TRUE),
    sd   = sd(flipper_length_mm, na.rm = TRUE),
    .groups = "drop"
  ) |>
  mutate(across(c(mean, sd), ~ round(.x, 1)))
```

3. A **paragraph that refers to both by their cross-reference**, e.g.
   *"Fig. @fig-flipper shows the spread of flipper length across species; the
   means in @tbl-flipper-summary confirm that Gentoo penguins have the longest
   flippers."* The `@fig-…`/`@tbl-…` numbers update automatically.

## Findings

Under `## Findings`, write a short, honest conclusion. Include what you did
**not** find. Example: *"Gentoo penguins have clearly longer flippers than
Adelie and Chinstrap; a formal test would be needed to say how significant the
gap is."*

## Done: check yourself

- Five headings: `The question`, `The data`, `Cleaning`, `Results`, `Findings`?
- Every **figure** and **table** has a **caption** *and* is **referred to in prose**?
- Does the report render from source with no manual steps?
- Would a stranger, opening only this `.qmd`, trust and reproduce it?

> **Stretch.** Add a second figure of `body_mass_g` by species, or a
> `## References` section with one citation (you'll need a `bibliography:`
> entry in the YAML header and a `references.bib` file).