Census data is often seen at a large scale — atlases, research studies and interactive visualizations all offer the view from 10,000 feet. But there are people inside those top-line numbers. And when you start to look at the people in the data sets, you get a glimpse of their lives. Just a few descriptors — how much they work, whom they take care of, where they were born — can give us a sense of the people around us.
Which is why I built a Twitter bot that mines for details in the data. Called censusAmericans, it tweets short biographies of Americans based on data they provided to the U.S. Census Bureau between 2009 and 2013. Using a small Python program, the bot reconstitutes numbers and codes from the data into mini-narratives. Once an hour, it turns a row of data into a real person.
I haven’t moved recently. I work for a private company. I was widowed.
— censusAmericans (@censusAmericans) July 24, 2015
I work less than 40 hrs a week. I went to college for less than a year. I’m on active duty for training in the Reserves/National Guard.
— censusAmericans (@censusAmericans) July 20, 2015
I have been married a few times, more than twice. I had less than 2 weeks off last year. I Drive to work.
— censusAmericans (@censusAmericans) July 15, 2015
CensusAmericans is a late addition to Twitter bots, which range from the creative to the utilitarian. I first discovered them through Allison Parrish’s everyword, which had a seven-year run of tweeting “every word in the English language” (109,000 tweets in all).
As Rob Dubbin pointed out in his New Yorker article on the genre, bots like Parrish’s excel at juxtaposing their automated tweets with a follower’s existing Twitter feed. They are often humorously, and sometimes illuminatingly, contextualized by headlines on one side and personal exchanges on the other.
CensusAmericans will insert strangers into your life at regular intervals and will continue its automated task until it gets to the end of the 15,450,265 rows in the data set. That’ll only take about 1,760 years.