Research & Insights

By Mike Love, April 1, 2008

Data Enrichment: A Power Law Distribution of Drunkenness

In the previous post I talked about how we scraped about one hundred photos from social networking sites and had Turkers perform data enrichment to guess those people’s traits – including age, race, intelligence, political affiliation, and intoxication.

Intoxication was listed as a checkbox. Averaging the guesses for each photo gives us an intoxication rating from 0 (sober) to 1 (smashed). In the histogram, you’ll notice the familiar power law distribution of drunkenness:

drunkogram.png

Finally, a cautionary graph. We found a clear correlation between perceived intoxication and perceived intelligence:

intelintox.png
Intelligence scale: 0 = “doofus”, 1 = “dull”, 2 = “average”, 3 = “bright”, 4 = “genius”

Mike