Opinions Still Divided Over Azerbaijani Exit Polls
By Khadija Ismayilova: 11/05/05

With less than a day before Azerbaijan’s parliamentary elections, unanswered questions still exist regarding the exit polls that were intended to serve as a check on official vote results. Among the more problematic issues, according to a Central Election Commission source: which of three main polls to use for comparison with official results.

One poll will be run by a group organized by the Washington, DC-based PA Consulting Group in 65 constituencies and funded by the United States Agency for International Development. New York-based Warren Mitofsky International and the Estonian firm Saar Poll will run polls in all 125 constituencies.

The source, who asked not to be named, said that the commission is quite confused by the number of polling companies conducting exit polls for the election, as well as by differences between their margins of error. Presidential administration officials have stated that an investigation of a polling station will be launched if the discrepancy between a polling station’s official result and that of an exit poll exceeds the standard margin of error of the exit pollster’s work.

The source said that the CEC for now has no set plan for which exit poll to use for comparison with official results. “We hope that there will be no discrepancies between the results of the exit poll companies,” the source said, adding that if such a discrepancy did occur “we will find a solution.”

Confusion also surrounds the margin of error which would be used as the benchmark for comparison with official results. Both Mitofsky and Saar Poll claim that their results have less than a 3 percent margin of error. PA Consulting Group’s website states that its polls have a 3.5 percent margin of error. However, during a visit to Baku earlier in October, Assistant US Secretary of State for Eurasia Daniel Fried told a press conference that the USAID- funded exit poll could have a margin of error of no more than 2 percent.

Speaking to journalists on October 27, Ali Hasanov, head of the presidential administration’s public and political affairs department, only said that three companies will conduct exit polls in Azerbaijan, and that all three are respected organizations.

In his May 11 decree on providing for free and fair parliamentary elections, President Ilham Aliyev ordered local authorities to provide all opportunities for conducting exit polls and “for [the] fair and transparent disclosure of the results of the parallel voting."

Unlike the government, the opposition has been outspoken in its reaction to the three exit pollsters. The opposition believes that both Mitofsky and Saar Poll have the government as their clients, albeit using front organizations. Reluctance by both Mitofsky and Saar Poll to discuss their clients has fueled these doubts. Mitofsky originally claimed that its exit poll is sponsored by a Switzerland-based company called Renaissance. Saar Poll has named its client as Santo Communications, a British financial institution, the company says. [For background see the EurasiaNet Insight archive].

On November 1, Turan News Agency reported that it had received a letter that allegedly identified the client of Mitofsky International as Renaissance Associates, a lobbyist firm which represents the Azerbaijani government in Washington. In response to the claim, Mitofsky International stated on November 2 that Renaissance Associates is a respected organization with which Mitofsky has had experience working.

Opposition accusations of bias have also dogged PA Consulting Group’s two local partners, Baku-based SORGU (Survey) and GORBI, an opinion research firm based in Tbilisi, Georgia, which will act as a technical advisor. “They have the reputation of being biased organizations,” said PFPA Deputy Chairman Fuad Mustafayev. “ We do trust PA Consulting, but not their contractors.” The opposition traces its reservations to a May 2005 opinion poll commissioned by GORBI and implemented by SORGU, which gave President Ilham Aliyev a 77-percent approval rating, and was incorrectly attributed to polling giant Gallup International.

Choice of constituencies is another trouble spot for poll critics. PA Consulting Group chose the 65 constituencies in which it will conduct exit polls by random selection. None of the constituencies include races between opposition leaders and nominees of the ruling Yeni Azerbaijan Party.

The head of one non-governmental organization who shares the skepticism about SORGU and GORBI, has commented that if the random selection procedure had taken place in the presence of journalists, belief in the poll’s objectivity would have been stronger. “Maybe it would be worthwhile to choose two organizations - one trusted by the opposition, another by the government,” said Leyla Yunus, director of the Peace and Democracy Institute.

Speaking to EurasiaNet earlier in October, PA Consulting Group representative in Baku David Hoffman said that the random selection of constituency is a widely used polling practice and that there is no reason to abandon it.

Domestic and international reactions to the polls and their role in the election process promise to be mixed. Elin Suleymanov, senior counselor in the presidential administration’s foreign relations department, commented that the exit poll can be viewed as an auxillary method for vote counting, but in no way official. Sabine Freizer, director of the International Crisis Group’s Caucasus Project, named international observers as a preferable means for determining how far and free the elections since they will see the actual process of voting.

For now, with many Azerbaijani voters, curiosity about the polls appears to be winning out over any doubts about their effectiveness. Baku resident Sona Mamedova, 18, will be voting in her first elections on November 6 and says she has thoroughly researched all three polling companies. Mamedova conceded that she has doubts about how effective the polls will prove as a check against official results, but added: “[I]t is good that we have an exit poll. . . I like the fact that the opinion of people like me is interesting to someone.”

Editor’s Note: Khadija Ismayilova is freelance journalist based in Baku.

Want more stories?
Go to the News & Views Archive

A voting booth at the Central Election Committee in Baku waits to be sent to a polling station. (Yigal Schleifer for Eurasianet)