Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
×
Twitter Social Networks

Twitter-Based Study Figures Out Saddest Spots In New York City 57

Nerval's Lobster writes "A new research paper from the New England Complex Systems Institute, titled "Sentiment in New York City" (PDF), attempts to pull off something that would have been impossible—or at least mind-bogglingly difficult and time-consuming—before the invention of online social networks: figure out the block-by-block happiness level of the biggest metropolis in the United States. In order to generate their 'sentiment map' of New York City, the researchers analyzed data from 603,954 Tweets (collected via Twitter's API) organized by census block. 'This method, combined with geotagging provided by users, enables us to gauge public sentiment on extremely fine-grained spatial and temporal scales,' read the paper's abstract. The study took emoticons and word choice into account when deciding whether particular Tweets were positive or negative in sentiment. According to that flood of geotagged Tweets, people are happiest near New York City's public parks, and unhappiest near transportation hubs. Happiness increased closer to Times Square, the declined around Penn Station, the Port Authority, and the entrance to the Midtown Tunnel. People were in a better mood at night and on weekends, and more negative about the world between the hours of 9 A.M. and 12 P.M. None of this is surprising: who wouldn't be happy amidst the greenery of a public park, or borderline-suicidal while stuck in traffic or waiting for a late train? The correlation between happiness and Times Square is almost certainly due to that neighborhood's massive influx of tourists, all of them Tweeting about their vacation. But as with previous public-sentiment studies, using Twitter as a primary data source also introduces some methodology issues: for example, a flood of happy Tweets from tourists could disguise a more subdued and longstanding misery among a neighborhood's residents, many of whom probably aren't tweeting every thirty seconds about a Broadway show or the quality of Guy Fieri's food."
This discussion has been archived. No new comments can be posted.

Twitter-Based Study Figures Out Saddest Spots In New York City

Comments Filter:
  • by Anonymous Coward

    Surprisingly, Ground Zero not the saddest place on Earth -- that's reserved for Wall Street.

    • Surprisingly, Ground Zero not the saddest place on Earth -- that's reserved for Wall Street.

      Unlikely, given that the only thing that Wall Street likes more than money is cocaine.

  • Not Tweeting. (Score:5, Insightful)

    by Lairdykinsmcgee ( 2500904 ) on Wednesday August 21, 2013 @09:14PM (#44638801)
    The saddest parts of New York City are not where people who own mobile devices and laptops convene. The saddest parts of New York City are where people are wearing trash bags, begging for food and shelter... They are not begging for attention by Tweeting their pretentious frivolity.
    • by Anonymous Coward

      Sad and unhappy is relative. I bet an alcoholic with no job, no family, and living off the street is much happier just to have bottle of vodka than someone just one block away getting home from work at 9:30pm to their 1+ million dollar condo. Someone in the middle looking at both people would naturally assume the later is happier.

    • Re:Not Tweeting. (Score:5, Interesting)

      by AHuxley ( 892839 ) on Wednesday August 21, 2013 @09:50PM (#44639049) Journal
      Yes would be interesting to see wage/wealth heat map over the happy map. Add in gov workers with jobs/pensions for life, rent control areas, crime and other data.
    • by Camael ( 1048726 ) on Wednesday August 21, 2013 @10:10PM (#44639187)

      The saddest parts of New York City are not where people who own mobile devices and laptops convene. The saddest parts of New York City are where people are wearing trash bags, begging for food and shelter... They are not begging for attention by Tweeting their pretentious frivolity.

      True. But I wonder how long will it be before the researchers apply the same techniques to analyse block by block the political beliefs of the residents. It may even be hyper accurate if you assume quote reasonably that :-

      1. those who tweet about their political beliefs tend to be more passionate about it and are more likely to vote; and
      2. those who don't are apathetic to politics and are less likely to vote.

      If they can gauge something as subjective as 'happiness', gauging something more definite like the voting predisposition of the residents of an area would appear to be a simpler task.

      • That is a HUGE set of assumptions. You would need to go do some real research to find out if that was remotely valid.

        Many people I know don't do social networking because it can have some pretty serious consequences and be used against you. They do vote though.

        • by c0lo ( 1497653 )

          Many people I know don't do social networking because it can have some pretty serious consequences and be used against you. They do vote though.

          Enjoy while it last. The time when voting can have pretty serious consequences and be used against you may be near.

        • by Camael ( 1048726 )

          That is a HUGE set of assumptions. You would need to go do some real research to find out if that was remotely valid.

          That's fair. Interestingly enough, there is such a study conducted by the Indiana University, Bloomington on the correlation between voting patterns and tweets [ssrn.com]. I'll skip to the findings here:-

          Is social media a valid indicator of political behavior? We answer this question using a random sample of 537,231,508 tweets from August 1 to November 1, 2010 and data from 406 competitive U.S. congressional elections provided
          by the Federal Election Commission. Our results show that the percentage of Republican-candid

      • True. But I wonder how long will it be before the researchers apply the same techniques to analyse block by block the political beliefs of the residents.

        You're behind the times. Political campaigns already do that.

    • Believe it or not, beggars make a lot of money. If they get a good spot or route, they'll make a ton more than most honest living. Think several hundred a day, no taxes. Of course, it's not easy to get or keep a good spot.

      The ones who are truly in need don't beg. They just are. Though usually, they're also perfectly happy being.

      The druggies and alcoholics who are homeless all got sent to Cali or down south. Winters are not as cold there.

    • Yeah the difference between a depressed dumpster diving hipster and a depressed dumpster diving hobo is the hobo isn't getting sent a $1K a week by their beverly hill parents to spend on designer faux opshop clothing.
      My guess is the hipster is actually pretty pleased with his ridiculous beard, and the hobo just can't afford a nice shave.

    • That's far from guaranteed. You'd be surprised how miserable people who have everything can be, while some people who have nothing can often laugh at their circumstances. Never judge happiness by material possessions.

  • False assumption (Score:5, Interesting)

    by DNS-and-BIND ( 461968 ) on Wednesday August 21, 2013 @09:20PM (#44638847) Homepage

    This assumes that everyone uses Twitter.

    It's funny, isn't it? People who are heavy users of Twitter are absolutely convinced that EVERYONE is on Twitter. Because in their tiny world, it's true. It's like when some World of Warcraft nerd starts spewing jargon in front of everyone and nobody knows what's going on...alts, tank, twitchy, whatever.

    What's really frightening is how data like this is being taken seriously. Stock markets move based on Twitter.

    • This assumes that everyone uses Twitter.

      Not necessarily .. if they're doing something equivalent to a poll where they can make predictions they might be able to paint trends.

      If they're just saying "wow, there's a really happy person here it must be a happy place", not so much.

      As you say, it's Twitter, and it definitely isn't representative of everybody. That doesn't mean that people don't pay attention to it for reasons I've never quite understood.

      • Not necessarily .. if they're doing something equivalent to a poll where they can make predictions

        To be valid such a poll has to be a random sample of the population. In this case that is simply not true: they are selecting a sample containing only rich twits. Those without the time, inclination or money to buy a suitable device to use twitter are excluded.

        • Remember the infamous headline "Dewey Defeats Truman"? How could the Chicago Daily Tribune have gotten it so wrong? Well, one reason was their polls that showed Dewey with an insurmontable lead. Polls that were conducted via telephone, back in the days when not everybody had their own phone. Turns out that people who didn't have telephones (i.e., poor people) voted very heavily for Truman.

    • by Anonymous Coward

      "twitchy"? PLUUESE! You mean "twink"? Uneducated masses should not express their opinions on important matters!

  • The third link (Score:4, Insightful)

    by HairyNevus ( 992803 ) <hairynevus&gmail,com> on Wednesday August 21, 2013 @10:06PM (#44639141)
    Does that guy really think he's tearing Fieri a new one by incessantly asking nothing but sarcastic questions for two pages of review? Seriously, learn some English writing and criticism techniques if you're going to be a critic for a living.
    • That article was actually very popular when it came out because it's entirely built of questions. If he had just slammed Fieri it wouldn't have gotten near as many views as it did, but because of the way it was written it appeared on sites that it never would have otherwise. It's marketing, not an English journal.

  • by kawabago ( 551139 ) on Wednesday August 21, 2013 @10:31PM (#44639297)
    NJ.
  • POS research (Score:1, Informative)

    by nicoleb_x ( 1571029 )

    I really hope everyone stops reading and reporting arbitrary but easy statistics that don't prove Jack.

  • It is nice when science brings us facts that looked obvious to crowd wisdom: nature is good for our moods, work is not. (yes, I know some of us have interesting jobs, but we are minority).
  • 2 things,

    1) I posted this yesterday afternoon [slashdot.org], which I fully understand. I shouldn't have expected to be selected. No big deal.

    But really?
    2) I've tweeted about enjoying Guy Fieri's food, on two [twitter.com] occasions [twitter.com]. It's actually not the worst thing ever: Fuck Pete Wells.
  • I'd like to colour code the UK in the same way using the "well-being" data found from this XLS file [ons.gov.uk] (look for "Average rating").

    Is there any way I can go about this efficiently? The software would need to recognize the locations (Aberdeenshire, Hampshire, Surrey etc.), ask me what column and range for the colour coding I want to use, and colour a map of the UK automatically. Does anyone know if any site (maybe an app from Google?) could do this?

    Here's another dataset [independent.co.uk] on UK nationwide happiness. It wo
  • that the happiest places are near green grass and trees, not concrete and blacktop.

    I would never have guessed people would be happiest when they're around some form of nature rather than jammed next to each other, having to hear every word the other person is saying or having to endure their antics.

    Shocking.

  • Why would the "New England Complex Systems Institute" choose New York City? Doesn't New England have it's own sad (or angry) place to study?
  • would be interesting if the research(ers) could remove the tourists. Maybe tourists are identified as folks who spent most of their time tweeting from a zip other than Manhattan (e.g. they tweet most of their time from Scranton, PA, so NYC looks pretty darn cool). Enough of us laypeople blathering on about twitter as a surrogate poling technique: Time for Nate Silver to ride in and call statistical BS on tweeting as relevant means of gauging sentiment. But maybe he's busy creating data for ESPN guys callin
  • I see! people who use twitter are sad, sad people...!

MS-DOS must die!

Working...