Polling in the new era of Italian politics

"Full of strange ideas: Beppo Grillo in Bologna." Photo by  antonella.beccaria via Flickr.

Full of strange ideas: Beppo Grillo in Bologna.” Photo by antonella.beccaria via Flickr.

In late Feb­ruary, some­thing hap­pened to the Italian gov­ern­ment that had never hap­pened before: a hung par­lia­ment. After 75 per­cent of the pop­u­la­tion turned out to vote, it took two days to tally the results. Now, almost three weeks later, the center right and center left par­ties remain in a stead­fast grid­lock. A third party–the so-​​called 5 Star Movement–received only mar­gin­ally fewer votes. Which wouldn’t be all that exciting, except that the 5 Star Move­ment is just a single guy, and that guy is a comedian–Beppe Grillo–who refuses to speak on tele­vi­sion or radio, lest his mes­sage be skewed. Instead, he blogs and he speaks in the piazzas around Italy.

The hung par­lia­ment and the come­dian con­tender mark an odd time for Italy, which will have it’s next elec­tion in less than a year, according to people who know about Italian pol­i­tics (of which I’m not one). Things are changing online and off, and the old ways of doing things are clearly not holding up. That’s even true of the polling methods that were used to pre­dict the out­come of the elec­tion in the weeks leading up to it.

In a great article on the Daily Beast, which gives an awe­somely read­able and acces­sible account of the ever-​​so-​​complicated Italian polit­ical cli­mate, Tim Parks recaps the final counts for the four main can­di­dates, Luigi Bersani, Mario Monti, Silvio Berlus­coni, and Grillo, respectively:

So 30 per­cent to grim sup­port of the old work­place; 10 per­cent to the gen­tleman upholding the inter­na­tional mon­e­tary system; 30 per­cent to the rich, old guy inviting us all to have a good time; and 25 per­cent to the wild man who wants to kick ass.

How did the poll­sters fair? Not great. They gave 40 per­cent to Bersani (the old work­place), 20 per­cent to Monti (the gen­tleman), less than 20 per­cent to Berlus­coni (the old rich guy), and 15 per­cent to come­dian Beppe.

North­eastern pro­fessor Alessandro Vespig­nani was also watching the elec­tion closely, and he and his team at the Lab­o­ra­tory for Mod­eling Bio­log­ical and Socio-​​technical Sys­tems made a few pre­dic­tions of their own. They got all of them right except for one: Berlusconi.

How did they do it? And why were they off for the old rich guy? Well, it all goes back to that old ways thing I said ear­lier. Calling people on the phone seems to no longer be the eas­iest or most effi­cient way to probe society’s sen­ti­ments. In col­lab­o­ra­tion with researchers at the Insti­tute for Sci­en­tific Inter­change Foun­da­tion in Italy, Vespignani’s team tracked voter’s inten­tions by what they said on Twitter, instead. And actu­ally, according to one of the project coor­di­na­tors, North­eastern researcher Nicola Perra, “We did not make pre­dic­tions. We just studied the raw signal, and there are biases, in geog­raphy, age, et cetera.”

On their web­site, Tweet­Politik (it’s all in Italian), the team demon­strates results from three dif­ferent types of analyses, which they per­formed in the weeks leading up to the election.

First, they made activity maps using geo-​​localization data. These show where in the country people tweeting about each of the can­di­dates reside. “Each dot in the maps is a tweet. The trans­parent cir­cles rep­re­sent how intense the signal was in each census area,” said Perra. So, the map of Twitter activity about Beppe Grillo looks like this:

Activity map displaying tweets mentioning 5 Start Movement candidate, Beppe Grillo. Image via TweetPolitik.

Activity map dis­playing tweets men­tioning 5 Start Move­ment can­di­date, Beppe Grillo. 53% of the tweets came from the northern region of the country. Image via TweetPolitik.

While the one for Berlus­coni looks like this:

Activity map showing tweets for Berlusconi. Image via TweetPolitik.

Activity map showing tweets for Berlus­coni. Nearly 50% of these tweets came from the North. Image via TweetPolitik.

Con­ver­sa­tion maps show the high­ways of con­ver­sa­tion on Twitter for each party. These data came from hash­tags asso­ci­ated with the var­ious par­ties and rep­re­sent people in dif­ferent areas tweeting about the same thing:

Conversation map showing Twitter discussions using hashtags associated with Bersani's party, the center left. Image via TweetPolitik.

Con­ver­sa­tion map showing Twitter dis­cus­sions using hash­tags asso­ci­ated with Bersani’s party, the center left. Most con­ver­sa­tions took place between res­i­dents of large met­ro­pol­itan areas, usu­ally in the North or South of the country. Image via TweetPolitik.

Finally, (and this one is my favorite), the mon­i­toring stream graphic shows how var­ious dis­cus­sion topics evolved over time in a ten day period sur­rounding the elec­tion.  On Feb­ruary 19th around 7 o’clock, Beppe was a pop­ular subject:

A snapshot of twitter activity by Italian voters in the days leading up to the election on February 24. Image via TweetPolitik.

A snap­shot of twitter activity by Italian voters in the days leading up to the elec­tion on Feb­ruary 24. Image via TweetPolitik.

All of this is just reporting the signal that they saw in the data. The “pre­dic­tions” (which weren’t really pre­dic­tions, but just sig­nals) came from the global share of activity seen for each party. This stuff isn’t pre­sented on the site.

In summa, some num­bers were really close to the final out­come, others were not,” said Perra. “The rea­sons of the dis­crep­an­cies are biases. We could have cor­rected the sig­nals con­sid­ering age dis­tri­b­u­tion, Twitter pen­e­tra­tion in dif­ferent areas, but we decided not to make it about pre­dic­tion, at this round.”

But this still leaves the ques­tion of why Berlus­coni evaded their glance, even if it wasn’t a pre­dic­tion. That has to do with social phe­nomena, said Vespig­nani. Berlus­coni promised to repeal an impor­tant but hefty tax put in place by the last pres­i­dent, Monti (who actu­ally took over for Berlus­coni when he was obvi­ously flailing during the height of the eco­nomic crisis). Everyone knew this was a bad idea for the country, but it was also incred­ibly tan­ta­lizing for each indi­vidual tax­payer. So, no one in their right mind would claim to sup­port Berlus­coni, in person or on Twitter. But behind the safe shield of a voting booth? That’s another story.