Research Resources

General

• Kagi — a paid ad-free search engine with bells and whistles
• Obsidian — take notes
• Bear — take notes
• Overleaf — collaborative LaTeX
• Mermaid — draw diagrams
• Airtable — spreadsheet/database mash-up
• Toggl — tracking your time
• Zapier — automating stuff
• Julia Evans on How to Ask Good Questions
• Trey Causey’s Do You Have Time for a Quick Chat?
• Chicago Booth Clark Center Panels — “Chicago Booth’s Kent A. Clark Center for Global Markets has assembled and regularly polls three diverse panels of expert economists…”
• VoxDevLits — “VoxDevLits are living literature reviews that summarise the evidence base on policy-relevant topics related to development economics in an accessible manner.”
• Ungated Research — “provide[s] access to all publicly available working papers for research in leading economics journals in one place.”

Coding and Data

• [book] R for Data Science 2e
• [book] Kieran Healy’s The Plain Person’s Guide to Plain Text Social Science
• Quarto – “An open-source scientific and technical publishing system”
• RStudio Desktop IDE
• Positron on GitHub IDE
• GitHub – version control
• GitLab – version control
• [paper] Karl Broman & Kara Woo on Data Organization in Spreadsheets
• [paper] Hadley Wickham’s Tidy Data
• Julia Evans’ Oh Shit, Git! zine

Methods and Stats

• [article] How are econometric methods applied by researchers in development economics? | VoxDev Blog
• [book] Ethan Bueno de Mesquita & Anthony Fowler’s Thinking Clearly with Data
• [book] Joshua Angrist & Jörn-Steffen Pischke’s Mastering ’Metrics
• [book] Joshua D. Angrist & Jörn-Steffen Pischke’s Mostly Harmless Econometrics
• [book] Aki Vehtari, Andrew Gelman, & Jennifer Hill’s Regression and Other Stories
• [paper] Andrew Gelman, Aki Vehtari and others on Bayesian Workflow
• [book] Jeffrey Wooldridge’s Introductory Econometrics: A Modern Approach
• [book] Nick Huntington-Klein’s The Effect
• [book] Scott Cunningham’s Causal Inference: The Mixtape
• [paper] Susan Athey & Guido W. Imbens’ Machine Learning Methods That Economists Should Know About
• [book] Dani Rodrik’s Economics Rules
• [paper] Angus Deaton & Nancy Cartwright’s Understanding and Misunderstanding Randomized Controlled Trials
• [book] Graeme Blair, Alexander Coppock, & Macartan Humphreys’ Research Design in the Social Sciences
• [paper] Sayash Kapoor and others REFORMS: Consensus-based Recommendations for Machine-learning-based Science
• [book] Chester Ismay and Albert Kim’s ModernDive Statistical Inference via Data Science in R
• [book] Geocomputation with R — “a book on geographic data analysis, visualization and modeling.” by Robin Lovelace, Jakub Nowosad and Jannes Muenchow. See also the geocompx project for resources in R, Python, and Julia
• [paper] Natalie Ayers, Gary King and others: Statistical Intuition Without Coding (or Teachers)
• Seeing Theory — interactive stats 101 visualizations
• Common statistical tests are linear models (or: how to teach stats) by Jonas Kristoffer Lindeløv
• IZA’s methods write-ups
• Evidence in Governance and Politics (EGAP) Methods Guides
• The World Bank’s Curated List on Technical Topics
• J-PAL’s Research Resources

Data Collection

• J-PAL’s resource on survey programming (small contribution by me)
• SurveyCTO see also the free Community Subscription
• J-PAL’s Repository of measurement and survey design resources
• Surveying Young Workseekers in South Africa (contribution by me) | Southern Africa Labour and Development Research Unit (SALDRU) blog

Dataviz

• From data to viz — “…leads you to the most appropriate graph for your data. It links to the code to build it and lists common caveats you should avoid.”
• [free] RawGraphs
• [free] Tableau (public)
• Datawrapper
• Flourish
• Data Visualization Society
• J-PAL’s resource on data visualization (small contribution by me)
• Data by Design — “An Interactive History of Data Visualization”
• Frank Elavsky’s Chartability “…is a set of heuristics (testable questions) for ensuring that data visualizations, systems, and interfaces are accessible”

LLMs

• See my post on LLMs for a general audience
• GPTs
  • Claude
  • ChatGPT
  • Meta AI’s llama-3 or chat here
  • Mistral
  • Gemma
  • WizardLM
• Some prompting patterns
• Google’s NotebookLM
• The Financial Times’ piece Generative AI exists because of the transformer
• Transformer Explainer — “an interactive visualization tool designed for non-experts to learn about Transformers through the GPT-2 model” by Aeree Cho, Grace C. Kim, Alexander Karpekov and others
• [video] Andrej Karpathy’s LLMs for busy people
• Ted Chiang’s ChatGPT is a blurry JPEG of the web
• [paper] Michael Townsen Hicks, James Humphries & Joe Slater’s ChatGPT is bullshit
• Worth reading re copyright law: The New York Times’ case against Microsoft and OpenAI
• Jaron Lanier’s How to Picture A.I.
• [paper] Murray Shanahan’s Talking About Large Language Models — careful of anthropomorphizing LLMs, or assigning intent, it may affect how successful you are at using them
• [paper] Emily Bender, Timnit Gebru et al. On the Dangers of Stochastic Parrots
• Simon Willison’s Prompt injection and jailbreaking are not the same thing — Simon discusses some of the ways in which LLMs can be vulnerable
• Rohit Krishnan’s What can LLMs never do?
• An interesting read on the process of fine-tuning an LLM’s “character” by Anthropic: Considerations in constructing Claude’s character. I think Patrick House’s article is a great complement: The Lifelike Illusions of A.I
• Simon Willison’s Think of language models like ChatGPT as a calculator for words — a nice metaphor for LLMs, see also “the weird intern”
• Simon Willison’s Embeddings: What they are and why they matter
• Vicki Boykis’s list of “no-hype” reads on LLMs — great readings on the fundamentals of LLMs and more
• [book] AI Snake Oil: What Artificial Intelligence Can Do, What It Can’t, and How to Tell the Difference by Arvind Narayanan and Sayash Kapoor
• LLMs in production: Hamel’s “Your AI product needs evals” — relevant for when you are starting to think about the effect of your prompt tweaks on your outputs, and whether you are making things better or worse
• The What We Learned from a Year of Building with LLMs series by Eugene Yan, Bryan Bischof and others
• Against LLM maximalism by Matthew Honnibal (Explosion, spaCy) — “Instead of throwing away everything we’ve learned about software design and asking the LLM to do the whole thing all at once, we can break up our problem into pieces, and treat the LLM as just another module in the system…”

NLP

• [book] Emil Hvitfeldt and Julia Silge’s Supervised Machine Learning for Text Analysis in R
• [book] Julia Silge and David Robinson’s Text Mining with R
• The STM R package for structural topic modelling
• The textnets R package
• The Quanteda R package for working with text data
• BERTopic — python package for topic modelling with embeddings
• See spaCy and some of their demo projects for stuff like text categorization or custom Named Entity Recognition (NER)

Courses

• Harvard’s CS50: Introduction to Computer Science
• Harvard’s CS50: Introduction to R
• Grant McDermott’s Data science for economists
• EconDL — “comprehensive resource detailing applications of Deep Learning in Economics.”

Writing

• Benjamin Dreyer’s Dreyer’s English: An Utterly Correct Guide to Clarity and Style — see also Katy Waldman’s review in the New Yorker The Hedonic Appeal of “Dreyer’s English”
• Verlyn Klinkenborg’s Several Short Sentences About Writing
• The Chicago Manual of Style, 18th Edition

Fitness

• [book] Casey Johnston’s “LIFTOFF: Couch to Barbell”
• Megan Gallagher’s “Stronger by the Day”
• Macrofactor — nutrition
• {…}