EVR 628- Intro to Environmental Data Science

Week 1: Introduction to Data Science and RStudio IDE

Juan Carlos Villaseñor-Derbez (JC)

Rules of the game

Course materials and content

Course website at: jcvdav.github.io/EVR_628

Grading

  • Assistance and participation: 50% - come prepared to participate
  • Assignments: 40% (n = 4; 10% each, graded as completed / not completed)
  • Final presentation / Data Science portfolio: 10%

The (general) instruction format

Lecture sessions (Tuesdays)

  • Slide-based delivery of content
  • Emphasis on the why of an approach

Hands-on sessions (Thursdays)

  • No slides, mostly live coding (you follow along)
  • Emphasis on how of an approach
  • Expect the first sessions to be rocky as you get up to speed and we standardize our computers

Important

Make sure to complete assigned readings or software installation before class. This will count towards participation.

Interactions with me

  • Office hours: Thursdays 10:15-12:00
  • Via email, please include EVR628 in your subject heading

Adaptive approach

  • About three weeks in I will poll you (anonymously)
  • Then again 5 weeks in (anonymously)
  • Speed things up? Slow things down?
  • Are materials relevant to you?

Quick notes on using these slides

  • Navigation bar on the left
  • Click and zoom on figures
  • f for full screen / Esc to exit full screen
  • o for overview
  • e for PDF export mode (can’t see code)
  • ? for other help
  • Navigate with keyboard
  • Clicking on links
Code 
plot(mpg ~ disp, data = mtcars)

About you

  • Preferred name
  • Year and program
  • Interests

Let’s jump in

Learning objectives

By the end of this week, you should be able to:

  • Open the RStudio IDE
  • Identify the main work areas
  • Start a project and a script
  • Install R packages

Introduction to Data Science

What is Data Science?

Code 
library(tidyverse)
library(ggrepel)

tibble(skill = c("Coding", "Stats", "Domain"),
       x = c(-1, 1, 0),
       y = c(1, 1, -1)) |> 
  ggplot(aes(x = x, y = y, label = skill, color = skill)) +
  geom_point(size = 100, alpha = 0.5) +
  geom_text(mapping = aes(label = skill),
            color = "black",
            nudge_x = c(-2, 2, 0),
            nudge_y = c(0.5, 0.5, -2)) +
  geom_text(x = 0, y = 0, label = "Data Science",
            color = "black") +
  coord_equal() +
  lims(x = c(-4, 4),
       y = c(-4, 4)) +
  theme_void() +
  theme(legend.position = "none")

Conceptual model of data science

This course only covers one bubble

How will we approach data science?

How we think about projects

Modified from Wickham, Mine Cetinkaya-Rundel, and Grolemund (2023)

How we’ll learn things

  • Organization
  • Visualization
  • Reproducibility
  • Data management, transformation, wrangling
  • Spatial data and other extensions
  • No modelling!
  • No communication(-ish)

A word of caution

  • This will feel weird
  • It will be painful

BUT!

  • It should feel weird
  • You are not alone; leverage resources:
    • Myself
    • Your TA
    • Each other
    • LLMs
  • It will be worth it
  • If we have to slow things down, we will

Use of LLMs

  • You are allowed to use LLMs to troubleshoot your code
  • I discourage you from asking your LLM to make the assignments for you
  • At the end of this course you should be literate enough in R to do some vibe coding
  • Becker et al. (2025):
    • 16 experienced software developers
    • 246 tasks assigned
    • How much time will you save using AI? (-24%)
    • AI randomized
    • How much do you think you saved using AI? (-20%)
    • Actual completion times with AI (+19%!)

Motivation to learn scripting languages

  • Reproducibility:Data and code are increasingly required by publishers
  • Efficiency and flexibility
  • Scale: From small to medium data, it’s all the same
  • Responsibility:
    • We spend thousands of hours and dollars collecting data
    • Objective is to understand how the world works / improve the state of the world
    • Bad practices can erode all this away

Our toolkit

  • R will be our engine for statistical computing
  • RStudio will be our Integrated Development Environment (IDE)
  • Git will help us keep track of our code
  • GitHub will allow us to share our code with the world
  • EVR628tools contains data and a few functions to support the course

You should have already done this

  • Download and install the latest version of R (currently v 4.5.x) for MacOS or Windows
  • Download and install the latest version of RStudio Desktop
  • Full set of instructions available here.

Why an IDE

Multiple platforms

  • Keep data in Excel (needs microsoft license)
  • Statistical analyses in some GUI (needs IBM SPSS license)
  • Build figures in Tableau (needs license)
  • Copy things to Powerpoint / Word (need….)
  • Ooops, need to update my data…

R Studio

  • Free and open-source
  • One single platform to:
    • Talk to R, Python, SQL, html….
    • Maintain data, analysis, figures, and even text
  • Cognitive costs of switching between screens
  • There are other solutions (posit, VSCode, Cursor, Codium…)

RStudio projects

  • RStudio projects keep all files associated with a project together
  • One folder = one project
  • In my case, 1 paper = 1 project | 1 class = 1 project
  • This entire course will be one project in your computer
  • A project tells R what the “working directory” is (where to find files)
  • Needs to live in your computer (not the cloud!1)

That’s it for today

(Unless you are having trouble installing R / RStudio)

Getting to know R Studio

Hands-on tasks:

References

Becker, Joel, Nate Rush, Elizabeth Barnes, and David Rein. 2025. “Measuring the Impact of Early-2025 AI on Experienced Open-Source Developer Productivity.” https://arxiv.org/abs/2507.09089.
Wickham, Hadley, Mine Cetinkaya-Rundel, and Garrett Grolemund. 2023. R for Data Science: Import, Tidy, Transform, Visualize, and Model Data. 2nd ed. Sebastopol, CA: O’Reilly Media.