Just who are these #OccupyGezi protesters?
Finance Minister Mehmet Şimşek responded to a tweet of mine on June 15, where I was lamenting that even voices of reason in the government, such as himself and economy tsar Ali Babacan, were resorting to conspiracy theories: “On numerous occasions I stated that I didn’t believe in conspiracy theories.”
I guess assertions such as international media attacking the Justice and Development Party (AKP), or the interest rate lobby trying to ruin the Turkish economy, do not count as conspiracy theories. But I was more interested in the minister’s next tweet: “However, there is plenty of evidence that events were organized and highly ideological.” I wanted to look into this claim using data.
Public opinion research and consultancy firm Konda interviewed 4,411 protesters in Gezi Park on June 6-7, and after publishing their results on June 13, they were kind enough to share their data with me. A first look hinted that the minister could be right, after all: Although 17 percent of them were not old enough to vote, only 2 percent stated they had voted for the AKP in the 2011 elections, when the party got 50 percent of the vote.
But if the protesters have a political affiliation, it is anti-AKP more than anything else. The 41 percent of respondents who voted in 2011 for the main opposition Republican People’s Party (CHP), which is accused by the AKP of being behind the protests, were much more than the party’s nationwide 26 percent support. However, only 31 percent of the surveyed said they would vote for the CHP today. 29 percent were undecided, and 18 percent would not vote.
Statistical analysis revealed that die-hard CHP voters, i.e. those who voted and would vote for the CHP, were only slightly more likely to be at Gezi as a reaction to the AKP and Prime Minister Recep Tayyip Erdoğan. Similarly, there weren’t any meaningful difference between the two-thirds of respondents who were regulars at the park and the remaining one-third who were first-timers or had been there once or twice before.
If Şimşek is intent on finding a phantom menace behind the protests, the best candidate would be the police. Half of those surveyed decided to go to the park after seeing the police violence against the protesters. Similarly, according to social media big data firm ebrandValue, tweets with hashtags about the protests shot up right after the first police crackdown.
I then tried to figure out why there were protests in certain cities. Using data on protest-related injuries from the Turkish Medical Association and hashtag figures by city from ebrandValue, I identified 14 cities where there were widespread protests and used econometrics to explain incidence of protests with two dozen city-level social and economic indicators.
Lo and behold, population, education and the AKP’s share of votes in 2011 turned out to be the only meaningful factors after a million regressions. So the protesters could be you... or me. They could be all of us, at least those of us who did not vote for the AKP.