FEATURE10 July 2009

Zogby on polling challenges in Albania

Europe News

Voters in Albania went to the polls on 28 June, and Zogby International was there for the first time to conduct exit polls.

Zogby’s poll gave the Democratic Party’s coalition, led by Prime Minister Sali Berisha, a 49% to 41% lead over the opposition Socialists, and the pollster predicted the Democrats would win 69 of the 140 seats in parliament. The final result of 47% to 45% (yet to be confirmed by a recount) gives the Democrats a projected 71 seats. Not a bad performance for a poll in a country where opinion research remains in its infancy.

Research spoke to John Zogby, president and CEO of Zogby International, and Sam Rodgers, who managed the Albanian polling project, about the challenges and lessons of polling in a young democracy.

What’s the significance of this project for Albania?

Zogby: What I’ve learned in polling difficult places, especially at election time, is that the process is as important as the content. If we can do the work, and do the work reliably and accurately, that in itself says a lot about democracy and readiness for reform. You’ve had tremendous change in a country like Albania in the sense that there is a freer market, there is a freer government and that breeds an atmosphere where people feel free to talk. It also means someone coming round with a clipboard asking you lots of questions is not necessarily a bad thing.

Rodgers: There were a lot of people who said – “You should be cautious because there’s going to be a lot of intimidation at the polls”. People told us all these stories about problems in previous elections. However, on election day things were relatively smooth. The final report is not yet out from the OSCE [the European Union’s election monitors], but we saw no evidence of harassment or intimidation. We were never in danger, we didn’t experience any of those problems. It’s a coming of age not only for Albania but for polling over there.

Zogby: This is the first time that detailed polling was undertaken in an Albanian election. In our dealings with [TV station] Top Channel, they were very nervous, because we no sooner announced our relationship with them than there were all sorts of charges levelled that Zogby was biased towards the Socialists or that Zogby was in the back pocket of the Democrats. To be perfectly honest with you, during the first several months I didn’t even know who was running and I didn’t particularly care who was going to win. I was in a perfect position to be non-biased.

You had good response rates. Why do you think that was?

Zogby: We find response rates in developing countries, burgeoning democracies, to be substantially higher than in the UK, or in particular the US. Here in the US a great election turnout is 56% of eligible voters. The responsibility to vote seems to be much more greatly appreciated in areas where it’s been limited.

You made a deliberate attempt to delve deeper than simply trying to discover who people were going to vote for. Why?

Zogby: You need to explain why someone’s leading, or not leading and why there is movement. That elicits a more honest response. If I go back to the Mexico election in 2000, Labastida was the representative of the PRI – a party that had been in power for 70 years. Our poll with Reuters was the first to show the challenger Vicente Fox ahead by a point or two. What we did was not only ask how people were going to vote, but we also asked questions like, “Does the PRI deserve another term? and “Is the country headed in the right direction?”. The peasantry indicated that they were going to vote for Labastida but when we asked if the PRI deserved another term, only 27 or 28 per cent said ‘yes’. And when we asked if the country was going in the right direction, only 29 per cent said ‘yes’. Our last poll had the two candidates pretty much tied, and our statement said, if this is an honest secret ballot then Fox will win this election. This is the same sort of thing we saw in Albania. A razor-thin election, but we saw the Democrats go from a five-to-seven-point deficit to a dead-heat even to a two-point lead.

How does the work you did compare to past polling in Albania?

Rodgers: This was the first time they had an election under this new system and only the second election that you could call ‘free and fair’ (although the OSCE didn’t call the last one free and fair). There had been several polls before but they were mostly small samples. We took a very different approach. We took it nationwide with a sample of 1000 – the same way we’d do it in the US.

There was nothing we could work from in terms of historical data, poll reports or methodologies. One of the things we’ve tried to do is put up the methodology and the data on the web, so that we can build that kind of record. Local census information is very hard to come by as well, and because they’ve changed the election system each time, there’s no election data to work from.

What also makes it more difficult is not having the historical record to help us gauge turnout. It’s hard for us even in the US to make predictions based on 200 years of records.

What can you learn from this experience and apply to polling in other countries?

Rodgers: Every country is unique when it comes to this kind of election work. One of the things I think I’ve learned is you’ve really got to have a good understanding of the historical context that this work is being conducted in and apply that. I was there for the week preceding the election and one of my colleagues was there in the months before that. I went over there early with that purpose, I met with a group of university students at the University of New York in Tirana and met up with some other local leaders and OSCE election monitors.

Do you think more foreign firms will head into Albania to conduct polling now?

Rodgers: Yes and that’s really where the future for it has to be. Different media, TV stations will compete with each other, and that competition is a good thing. The next time every TV station and newspaper will probably be doing it if they can afford to.