All frameworks and cheat sheets

Benchmark range

How many Americans are unique on ZIP, gender and birth date

Range chart comparing two census studies: 87% vs 63.3% of Americans unique on ZIP, gender and birth date, and 18% vs 14.8% at county level.
Even the lower, better-documented figure means about three in five people are unique Source: Latanya Sweeney, Carnegie Mellon, 2000, and Philippe Golle, 2006

Two studies measured how many people in the US could be singled out by gender, location and full date of birth alone. Latanya Sweeney found 87% on 1990 census data at 5-digit ZIP level and 18% at county level. Philippe Golle, working with the 2000 census, found 63.3% and 14.8%. The gap is unexplained, so cite the smaller number, which still shows how three ordinary fields can identify most people.

Where this appears

What's in this benchmark range

Sweeney, 1990 censusGolle, 2000 censusRange
5-digit ZIP, gender, birth date87%63.3%63.3% to 87%
County, gender, birth date18%14.8%14.8% to 18%

Golle could not explain the gap. His re-run on 1990 census data found 61%.

Use this image

Free to use in your docs, decks and posts under CC BY 4.0, as long as you credit Koji with a link to this page. To embed it, paste this into your page:

html
<a href="https://www.koji.so/visuals/zip-gender-birth-date-uniqueness-sweeney-vs-golle"><img src="https://www.koji.so/visuals/zip-gender-birth-date-uniqueness-sweeney-vs-golle/zip-gender-birth-date-uniqueness-sweeney-vs-golle.png" alt="Range chart comparing two census studies: 87% vs 63.3% of Americans unique on ZIP, gender and birth date, and 18% vs 14.8% at county level." width="1200" height="630" /></a>
<p>Source: <a href="https://www.koji.so/visuals/zip-gender-birth-date-uniqueness-sweeney-vs-golle">How many Americans are unique on ZIP, gender and birth date</a>, Koji</p>