Final Project

Welcome to the final project! This is your chance to bring together everything you’ve learned — data wrangling, visualization, R programming, and reproducible research — into a single, polished analysis.

• Your Mission: Perform a complete data analysis and produce a fully reproducible Quarto report with publication-quality figures.
• You’ll submit:
  • A zipped RStudio project folder containing your Quarto document, data, bibliography, and all rendered outputs.
  • See Step 1 for the exact folder structure.
• Deadline: June 7th, 2026 — 11:59 PM

How You’ll Be Graded

Criterion	Weight	What we’re looking for
Required elements	40%	Does your report include every component listed below (YAML, introduction, analysis, visualization, references)?
Readability	15%	Are axis labels, titles, and annotations large enough to read comfortably? Is the narrative clear and well-organized?
Aesthetics & creativity	15%	Did you go beyond the minimum? Custom themes, smart color palettes, clean layouts, thoughtful narrative, etc.
Analysis depth	20%	Is the analysis substantive? Did you use appropriate wrangling, transformations, and/or programming techniques from the course?
Reproducibility	10%	Does your project render top-to-bottom from the zipped folder without errors? Are file paths relative?

So, get your creativity (and code) fired up. Let’s begin!

Step 1 — Project Setup

1. Create a new RStudio project for this assignment.
2. Organize your files — your project folder should follow this structure:

Helpful tools — project structure

final_project/
├── final_project.Rproj
├── data/
│   └── my_data.csv
├── final_project.qmd
├── references.bib
├── final_project.html    ← rendered output
└── final_project.pdf     ← rendered output

Use here::here() for all file paths so your project works on any machine (see Chapter 9):

library(here)
my_data <- read_csv(here("data", "my_data.csv"))

Step 2 — Choose Your Data

Pick one of the following options:

1. Option A: Analyze Your Own Data — Use a dataset you have collected or found online (from a reputable source). The dataset should not be one we used directly in class examples.

2. Option B: Reproduce/Extend a Published Analysis — Find a published research paper that includes data analysis. Reproduce (as closely as possible) some of its key results, and then extend or improve upon the original analysis or figures.

Helpful tools — finding data

If you don’t have your own data, here are some great sources:

• Dryad — open ecological and evolutionary datasets
• GBIF — global biodiversity records
• TidyTuesday — curated weekly datasets for R practice
• data.gov — U.S. government open data
• Kaggle Datasets — a wide variety of datasets across domains

Pick something you find genuinely interesting — the project is more fun (and better) when you care about the data!

Step 3 — Write the Quarto Report

Your final_project.qmd file must include the following sections. Think of it as a mini research paper:

YAML Header

Your Quarto document should begin with a proper YAML header:

Helpful tools — YAML setup

---
title: "Your Descriptive Project Title"
author: "Your Name, UCLA"
date: today
format:
  html:
    theme: cosmo
    code-fold: true
  pdf:
    documentclass: article
bibliography: references.bib
---

The code-fold: true option lets readers expand code blocks on demand, keeping the narrative clean (see Chapter 17).

Introduction

Provide a brief introduction that explains:

• The purpose of your analysis and the research question(s) you are addressing.
• The dataset you are using — where it comes from, what it contains, and why it’s interesting.
• A roadmap of what the reader will find in the rest of the report.

Data Analysis

Conduct a substantive analysis to address your research question(s). Your analysis should demonstrate skills from the course. This could involve (but is not limited to):

1. Data wrangling — importing, cleaning, joining, and transforming data using dplyr and tidyr (Chapters 7–9).
2. Descriptive statistics — calculating summary statistics, group comparisons, etc.
3. R programming — using vectorization, custom functions, or purrr to automate repetitive tasks (Chapters 19–22).

Important

Do not just show code and output — interpret your results. Explain what each analysis step reveals about your research question. The narrative is just as important as the code.

Visualization

Include at least two publication-quality figures. Each figure must include:

1. A descriptive title that communicates the main message (use labs(title = ...)).
2. Readable text sizes — axis labels, titles, and annotations should be large enough to read comfortably. The defaults are almost always too small (Chapter 6).
3. A non-default theme — pick one from jtools, hrbrthemes, ggthemr, etc. (Chapter 6).
4. Export with ggsave() — specify explicit width and height (Chapter 6).

Helpful tools — figure best practices

Remember the key lessons from visualization chapters:

• Match plot type to data — amounts (Ch. 10), distributions (Ch. 11), trends (Ch. 12), associations (Ch. 13), or other types (Ch. 14).
• Use color intentionally — pick a curated palette from MoMAColors, MetBrewer, PNWColors, etc. (Chapter 15).
• Advanced techniques — consider multi-panel layouts with patchwork, interactive plots with plotly, or annotations for emphasis (Chapter 16).

References

Cite at least two relevant sources using a BibTeX file (references.bib). You can use any citation style.

Helpful tools — managing citations

Use the references.bib file to store your citations. In your Quarto document, cite them with @key or [@key] syntax (see Chapter 18):

According to @wickham2016, tidy data has a specific structure...

You can find BibTeX entries on Google Scholar (click the cite button → BibTeX) or from journal websites.

Step 4 — Render and Submit

1. Render your Quarto document to both HTML and PDF formats.
2. Verify reproducibility — close RStudio, reopen the project, and render again from scratch. Everything should work without errors.
3. Zip the entire project folder and submit.

Reproducibility Checklist

Before submitting, make sure:

• All file paths are relative (e.g., "data/my_data.csv", not "/Users/YourName/Documents/...").
• All required packages are loaded with library() at the top of your .qmd file.
• No files are referenced outside the project folder.
• Both HTML and PDF outputs render without errors.
• Your code is well-commented, readable, and organized with meaningful variable names.