How Americans use Twitter to talk about flooding on the east coast?

Mark Graham

data show that

interactive version of the map

• More Data journalism and data The Guardian visualizations

Tamer Adham

My colleagues, Wang Ning, Scott Hale and I collect tweets containing the words "flood" and "flood" to examine how the use of twitter in the context of Hurricane Sandy could reflect experiences. In other words, we look at the shadows of human data and social event innate physical / material to see what they say.

Our original intention was also to assign references to floods in English and Spanish, to see if there are large geographical differences in language and social media reactions to the hurricane. With the rise of twitter crisis mapping and analysis, we thought it would be important to note the differences between English and Spanish (Spanish is the mother tongue of millions of people on the East Coast of the United States States.).

Second, we see that the data become less useful if you want to get ideas on a finer scale than the county. The data is good to reflect the broad path of the hurricane, but may be less useful for more detailed ideas. For example, it is unclear whether the number of tweets to gather in New York, compared to other places, which reflects the magnitude of the devastation of the city or just to say that New Yorkers are more likely to tweet about an event.

More importantly only won five tweets in Spanish during the same period! In other words, it is the absence of this map that are almost more interesting than the results assigned. The lack of content published in Spanish only English is necessarily includes content displayed in these representations. Absences in the rest of the country are also revealing.

Why do so few people in Kentucky, Missouri, Wisconsin, Twitter, etc. on the east coast floods? Is it because the act of tweeter about this event is really likely to be done by people on the ground, the experience of the storm? People are on the path of hurricanes simply not interested in the event? Or should we keep trying to make inferences from Twitter data that recognition of general patterns left by major events in the digital landscape?

Search

Mark Graham focuses on geographies of the Internet and the information in the Oxford Internet Institute, and overlap between ICT and economic development . It is also one of the creators of sheep floating blog

More data

More Data journalism and data visualisations The Guardian

government data

Development and aid data

• Find global development data with our gateway world
Can you do something with this data?

Find best price for : --Hurricane----Sandy----Twitter----Hale----Scott----Adham--