Around these parts, we know better than to compare the Upper and Lower Haights. With residents of both neighborhoods quick to defend their home turf, making assertions about which area is happier or safer can be a perilous endeavor.
But that's exactly what one brave man has done. And wisely, he did it with a slew of data backing him up.
Jesper Andersen, co-founder of a data visualization company called Bloom Studios, recently gave a talk at a conference in which he presented on the topic, "Building Data Narrative: Discovering Haight Street." Andersen was exploring whether he could pull data from sources like Twitter, Foursquare, and Google Maps to reveal a useful "description" of Haight Street -- and a scientific comparison of the two distinct neighborhoods. According to ReadWriteWeb, Andersen analyzed Foursquare checkins, crime data from DataSF.org, images from Google Street View, languages and topics from Twitter, and pictures from Instagram, mapping them along the length of Haight to tell the story of the street. So what did he discover?
For one thing, the Upper Haight is safer. In this slide, Andersen shows a line on the bottom depicting the elevation of Haight Street from west to east, and a line on top showing crime incidents from DataSF.org. There's a decisive mound of crime overlaying the Lower Haight (though, to be fair, we don't know the time period or type of crime statistics covered in this slide).
Slideshare/jesperandersen
Andersen also examined the languages people were using in their geotagged tweets along Haight Street. It turns out the Upper Haight featured more Swedish and Italian tweeters -- perhaps European tourists? -- while the Lower Haight had a large Thai representation.
Slideshare/jesperandersen
Finally, Andersen tried to get a sense of the overall attitudes of the two neighborhoods. Using "sentiment analysis" of tweets, he determined that in the Upper Haight, positive tweets significantly outnumbered the negatives, while the reverse was true in the Lower Haight.
Slideshare/jesperandersen
ReadWriteWeb notes the limitations of using data like this to draw too strong of a conclusion.
Jesper Andersen, co-founder of a data visualization company called Bloom Studios, recently gave a talk at a conference in which he presented on the topic, "Building Data Narrative: Discovering Haight Street." Andersen was exploring whether he could pull data from sources like Twitter, Foursquare, and Google Maps to reveal a useful "description" of Haight Street -- and a scientific comparison of the two distinct neighborhoods. According to ReadWriteWeb, Andersen analyzed Foursquare checkins, crime data from DataSF.org, images from Google Street View, languages and topics from Twitter, and pictures from Instagram, mapping them along the length of Haight to tell the story of the street. So what did he discover?
For one thing, the Upper Haight is safer. In this slide, Andersen shows a line on the bottom depicting the elevation of Haight Street from west to east, and a line on top showing crime incidents from DataSF.org. There's a decisive mound of crime overlaying the Lower Haight (though, to be fair, we don't know the time period or type of crime statistics covered in this slide).
Slideshare/jesperandersen
Slideshare/jesperandersen
Slideshare/jesperandersen
"It's not a perfect picture, but it's a start towards storytelling with data instead of just a jumble of numbers and facts that don't paint much of a picture at all."And the story it apparently tells is that we're a grumpy, dangerous, Thai-speaking bunch down here in the lower elevations. How does this analysis stack up with your assessment of the two 'hoods? (We have our opinion, but we're no fools -- we're keeping it to ourselves.)
