Home About Us Twitter Sentiment Analysis 2014 Turkey Presidential Election
2014 Turkey Presidential Election

We here at Monospect were curious about the Turkish presidential candidates and the impact of sentiment they had on Twitter; So we sought out to track that information between 4/8/14 - 7/8/14. Over the three month period, we tracked 51,566 tweets and the key points are highlighted below:

  • Out of the 51,566 tweets, close to 75% of the data is related to Recep Tayyip Erdoğan, Ekmelededdin İhsanoğlu in a close second, and finally Selahattin Demirtaş (with the least amount of "buzz" on Twitter).
  • As you can see on the chart provided, Ekmeleddin İhsanoğlu had the biggest variance of negative and positive tweets correlating to his name. Although Erdoğan won the election, he had a semi-small variance between positive/negative tweets. Candidate Demirtaş has almost equal parts of negative and positive tweets.
  • Overall, the positive-related tweets outweighed the negative for all three candidates. The negative rate is calculated bu looking at the ratio of all negative tweets per candidate.

In summary: This was an interesting study as one can now expand and correlate on how much of an impact social media can have on a campaign. More interestingly, the number of politicians overall are seeing the importance of social media and how quickly information can be obtained or distrusted onto these platforms. To dig a bit deeper, Recep Tayyip Erdoğan won the election in the first round by majority vote; He also had the highest number of tweets. It is obvious that he gained much more attention (at least through Twitter) thus becoming more popular to Twitter's feed and gaining more visibility with the public.

Note: The information we gathered from Twitter was completed automatically and without any interaction by our system. As stated, we reviewed over 50,000 tweets between April 8, 2014 and June 8, 2014. The nature of our questionnaire did not interfere with the collection of the data and the accuracy rate of this method is 70%.