Google Opens Election Survey Data to Crowdsourcing

Google wants data scientists and others to take a crack at its election survey results.

On Election Day, as millions of people headed to the polls to vote, Google’s Consumer Surveys team pushed out a survey with two simple questions:

  1. Who do you want to win the U.S. Presidential Election?
  2. Regardless of who you want, who do you think will actually win the election?

Google managed to collect some 10,000 samples between 2 AM PST and 11 PM PST, or the equivalent of one data point every 10 seconds. “Now that we’ve collected the dataset, we want to provide it to the community to see what interesting analyses, insights, and visualizations you can discover,” reads a note on the Consumer Surveys team’s Webpage. “This is a quest for the data scientists, engineers, statisticians, and political scientists out here.”

It’s a contest of sorts: whomever submits what Google views as the “most insightful” visualization or analysis of data will earn “one coupon towards free research using Google Consumer Surveys,” which isn’t exactly a new Wii U, but could interest those who work with datasets on a regular basis. The deadline for submissions is Dec. 20; the actual results of the survey are available here.

Contests aside, companies throwing their data to the masses for crowdsourcing could become the next big trend in Big Data. In October, for example, Greenplum (a subsidiary of EMC) announced it would integrate its Chorus analytics platform with startup Kaggle’s community of 55,000 data scientists.

Kaggle bills itself as a Website that makes “data science a sport.” It lets organizations post data problems to an extensive community of data scientists, who can provide an effective solution in exchange for a prize. “Those who are part of Kaggle’s community can choose to opt-in to doing contract work through Chorus,” read Greenplum’s Oct. 23 release on the integration. “From within the Chorus interface, Chorus users wishing to engage the Kaggle community will search, browse, and drill into profiles of Kaggle community members who are interested in collaborating together.”

Nor is Greenplum the only company in the crowdsourcing game. Whether a Washington, D.C.-based bike-rental company wanting a collective opinion on the best place to set up locations for new bikesharing stations, a human-rights group looking to document violence in Syria, or a Website devoted to hyper-local news, a wide variety of startups are embracing crowdsourcing as the ideal way to populate maps with all sorts of data-points—a way of visualizing information that would otherwise take many employees weeks (if not months or years) of painstaking work.

That doesn’t mean crowdsourcing is the way of the future—a lot of company data is far too proprietary to simply dump it in the collective hands of thousands of people. But as a way of gaining further insight (or at least creating some neat visualizations) it could become a more prominent arrow in your average company’s quiver.


Image: Google