themains / pwned Goto Github PK

We merge data from a large representative sample from YouGov (n = 5,000) with data from Have I Been Pwned (HIBP) to estimate how often people's information is exposed. And how that varies by race, education, income, and age.

We find that at least 82.8% of Americans have had their accounts breached at least once. And that on average Americans' accounts have been breached at least thrice. Better educated, the middle-aged, women, and Whites and Blacks are more likely to have had their accounts breached than the complementary groups.

• Data

  • YG Profile Data (CSV)
    • YG Profile codebook (pdf)
  • HIBP Data on the people (CSV)
    • HIBP v2 API (pdf)
    • HIBP codebook (xlsx)
  • HIBP Data on Breaches (JSON)
  • Current Population Survey Data (xlsx) and CSV

• Analysis

  • Jupyter notebook
  • R script for fancy tables, plots, and replication
  • R Script for comparing YG to CPS

• Figs

• Tables

• ms: pdf, bib, and tex

Ken Cor and Gaurav Sood