Zülfikar Doğan
Jun 20 2018

Stark contrasts in leading pollsters’ estimations ahead of elections

Turkish people will go the to polls on June 24 to elect the new members of the parliament and a new president. According to the new system, the political parties can form alliances for the parliamentary elections. The governing Justice and Development Party (AKP) established the Popular Alliance with far-right Nationalist Movement Party (MHP) and Great Unity Party (BBP). The main opposition People’s Republican Party (CHP) is in alliance with newly established right-wing Good Party, Islamist Felicity Party, and central-right Democrat Party, calling themselves the Nation Alliance. The mainly-Kurdish People’s Democratic Party (HDP) is not in an alliance, however will to a large extent determine the results of the election, as the distribution of the seats in the Parliament will change profoundly in favor of the opposition if HDP succeeds in passing the 10 percent election threshold.

The Popular Alliance’s presidential candidate is the incumbent Recep Tayyip Erdoğan, while all parties in the Nation Alliance are putting up their own candidates; CHP with Muharrem İnce, the Good Party with Meral Akşener, and Felicity Party with Temel Karamollaoğlu. Selahattin Demirtaş, the former co-chair of HDP, who has been jailed since 2016 is the presidential candidate of his party. Doğu Perinçek, the leader of left-ultra nationalist Vatan Party, is the fifth candidate.

To win the presidency, a candidate should receive 51 percent of the votes in the election. If none of the candidates succeed in reaching such a score, a second round will be held on July 8 and the two candidates with the highest number of votes will compete. 

Since the beginning of the election campaign, the results of surveys conducted by various pollsters have been widely discussed, as they are accused of being unreliable. There are various reasons for suspicions about the results announced by the pollsters. First of all, many pollsters are organically or indirectly affiliated to political parties, which put their motivations into question. Secondly, this will be the first election under the new system approved in the 2017 referendum and, therefore, given the fact that the pollsters in Turkey provide little information about their methodology, some believe that their simulations based on the old system may not give reliable results. Thirdly, many believe that Turkish people refrain from revealing their true preferences to pollsters, given the atmosphere of repression and fear in the country.

In such an environment, there are striking differences between the estimations of two well known pollsters – Metropoll and Konsensus – who conducted their surveys on June 14 and 16.

According to Metropoll’s latest survey, conducted on June 14, ten days before the election, the expected scores of the presidential candidates in the first round is as follows – numbers in brackets show the results form the same company’s May survey: Recep Tayyip Erdoğan 46.2 percent (45.9), Muharrem İnce 24.5 percent (20.1), Selahattin Demirtaş 11.3 percent (11.4), Meral Akşener 9.2 percent (11.2), Temel Karamollaoğlu 2.2 percent (1.7), and Perinçek 0.4 percent.

The share of undecided voters was 3.9 percent in Metropoll’s survey, while 0.7 percent said that they would not vote for any of the candidates, 0.3 percent would cast protest votes, and 0.3 percent would not participate the election. 

While the CHP candidate İnce appears to have increased his votes by 5 points since May, there is a 2 point decrease in Akşener’s votes. Erdoğan’s votes increased by 0.3 points and Temel Karamollaoğlu, the candidate of the Felicity Party increased his votes by 0.5 points.

In the parliamentary election, before the undecided votes are distributed, the Popular Alliance gets 47.5 percent, while the Nation Alliance garners the support of 35.5 percent and HDP passes the election threshold with 11.5 percent.

When the undecided votes are distributed, the results change as follows: Popular Alliance 49.4 percent, Nation Alliance 38.1 percent, and the HDP 12.3 percent. The other parties will receive 0.2 percent of the votes.

According to Konsensus, president Erdoğan will declare his victory during the first round of the presidential election, while the People’s Alliance will receive 53–55 percent of the votes and the HDP’s chances of passing the election threshold is uncertain. The company’s general director Murat Sarı told Ahval that their survey was conducted on June 16. “Due to the limitations set by the Supreme Electoral Council, we cannot share our results with the public, but according to our findings, president Erdoğan wins the presidency at the first round. Muharrem İnce’s votes is significantly higher than his party CHP, with 33–34 percent. The most important factor determining the outcome is the significant fall in the votes of Meral Akşener, compared to the survey we conducted in previous months,” Sarı said.

“Moreover, the CHP candidates’ votes come from different segments. İnce not only garners the support of CHP voters, but also gets votes from Good Party and the Felicity Party supporters. This is why his votes are increasing. However, according to our findings, Erdoğan will easily win the election on June 24,” he added. 

According to Murat Sarı, the Popular Alliance will get 53–55 percent of the votes in the parliamentary election, while the Nation Alliance’s votes fluctuate between 35–40 percent. “In our survey, HDP fails to pass the election threshold with the votes to be cast inside Turkey. The party is at a critical point, may or may not pass the threshold. However, the AKP and HDP receive the highest number of votes from Turkish voters living abroad. Therefore we estimate that HDP will pass the threshold thanks to voters abroad, our findings indicate that,” Sarı said. 

Those striking differences between the findings of the Turkey’s two most credible pollsters demonstrate the difficulties in estimating the results of the election and analysing the voters’ responses. 

In the parliamentary elections, there is a difference of 7–9 points in the vote of the Popular Alliance when we compare the results of Metropoll and Konsensus, and there is a serious discrepancy between the two companies’ samples and the methodologies they use in measurement. 

While, Konsensus’s general director Sarı underlines that president Erdoğan will easily win the presidential election on the first round, according to Metropoll’s survey, Erdoğan will get 46.2 percent of the votes at the first round and he needs approximately 5 percent additional votes to claim victory with 50+1 percent. Moreover, there is a 10 point difference in the CHP candidate İnce’s vote in the two surveys.

In Metropoll’s survey, HDP easily passes the election threshold with 12.3 percent and the total share of the opposition votes will reach 50.4 percent, meaning that opposition parties will have the majority in the parliament. 

According to the results of Konsensus, the ruling AKP and its People’s Alliance will succeed in having majority in the parliament, while the HDP will barely pass the election threshold with votes from abroad. If Konsensus's survey results are correct, the Popular Alliance will get more than 400 seats in the parliament, meaning that it will have enough members to approve further changes in the Turkish Constitution.

The opinions expressed in this column are those of the author and do not necessarily reflect those of Ahval.