Resources

A running list of things I point people to, sorted roughly by stage: kids and families, students, researchers, and anyone starting a career in data. Nearly all of them are free.

For Kids and Families

For curious kids and the grown-ups exploring with them.

  • NASA Space Place: space and Earth science for kids from the National Aeronautics and Space Administration (NASA), as games, crafts, and short reads.
  • NASA Citizen Science: contribute to NASA research, from spotting exoplanets to tracking how humans can thrive in space.
  • SciStarter: citizen-science projects anyone can join (see also Zooniverse).
  • NASA STEM Engagement: science, technology, engineering, and mathematics (STEM) challenges and the NASA EXPRESS newsletter.
  • Scratch: a free, visual first language for programming, with Code.org for guided lessons.

For Teens and Undergraduates

Starting out in statistics, data, or programming.

For Graduate Students and Early-Career Researchers

Learning R and Stata, keeping your work reproducible, and finding your footing in research.

Starting a Career in Data Science

When I moved from internships into my first data scientist job, I shared what had worked for me with Dataquest and in a guest post for Forbes, and I still give the same advice.

  • Build the foundation first. Free courses go a long way: the two introductory computer science courses above, the MIT Statistics and Data Science program, and Andrew Ng’s Machine Learning Specialization. Learn the command line and Git early.
  • Take internships and research positions seriously. My first full-time role came from an internship where I worked to exceed what was asked.
  • Keep a small portfolio. Four projects that show depth and range are enough: clean, commented, version-controlled code; a written explanation of each; at least one that tells a story with data; and one you deployed, such as an application programming interface (API), a Shiny app, or a dashboard. Leave out routine homework.
  • Put your work where people can find it. A personal website and a GitHub profile do most of the work. GitHub Pages hosts a site for free, and Julia Silge’s site is a good model.
  • Apply before you feel ready. If you meet about half of a posting’s requirements, apply.
  • Search through people first. In order: recruiters, then friends, family, and colleagues, then career fairs, then general job boards, company websites, and tech job boards last. When you contact someone, lead with how you could help their team.
  • Work on communication as hard as on code. Analysis matters only as far as others can understand it and act on it.
  • Dataquest Data Science Career Guide: the longer guide my interview appears in, covering roles, resumes, portfolios, and interviews.

Data Privacy and Confidentiality

A few favorites from the resource list I compiled for the American Statistical Association (ASA) Privacy and Confidentiality Interest Group, where the full list of articles, books, and R and Python packages lives.

Open Science and Research Software

Tools and communities for building work others can check and reuse.

  • Posit: the RStudio integrated development environment (IDE), Quarto, and the tools around them.
  • rOpenSci: peer-reviewed, community-built research software (and its sibling pyOpenSci).
  • R-Ladies: a worldwide network of local R meetups, study groups, and workshops.
  • forkflow: my own R package and open-access book for specification and vibration-of-effects analyses.
  • Chris Albon’s notes: a big, searchable set of Python, machine-learning, and statistics snippets.