STEM's Jan Hartl - fascinated by changing trends in Czech society

Jan Hartl

2006 is a crucial year for the Czech Republic. In June the country's voters go to the polls in a much-anticipated clash between the ruling Social Democrats and the right-of-centre opposition Civic Democrats, with the Communists lurking in the wings. Who will people vote for? Will they bother voting at all? Are they happy with their lives? Do they still support their country's membership of the European Union? Do they want to ditch the Czech crown and adopt the euro? Analysing social, economic and political trends in Czech society is the work of the many polling agencies which have sprung up since the fall of Communism. One of the most successful is the private agency STEM. For this week's One on One, Rob Cameron speaks to STEM's director Jan Hartl about how the polls are produced, and how much we should rely on them.

Can you sum up exactly what STEM is and how it goes about its work?

"STEM is a private, independent organisation which focuses on sociological research and organises surveys. Our main subjects which we study are political issues, social problems and communication."

I'm interested in the science of opinion polls. How exactly do you collect data?

"This is a methodology which is exactly the same all over the world. It's based on creating representative samples of our population. There are various methods how to achieve that, and by running representative surveys we try to assess public opinion of various issues of daily life."

And how accurate are those surveys?

"Well, STEM runs the largest surveys in our country, regularly over 1,500 respondents, which is relatively large even in international terms, and the accuracy is usually plus minus two percentage points."

So fairly accurate but can you really generalise about, for example, the voting intentions of the nation's electorate by asking a few people on Wenceslas Square which way they're going to vote?

"It's not exactly a few people on Wenceslas Square. If you ask a few people on Wenceslas Square you just get interpretations of people which they connect with their behaviour, but this is not an accurate picture of public opinion. To do so you would need to have at least a few hundred respondents. The usual number of people studied all over the world is around 1,000 respondents, that's a kind of general norm."

And I suppose it's very important to ask not just the people on Wenceslas Square but the people in little villages in South Moravia and in industrial areas of North Moravia and so on.

"Yes, that's exactly the case. You have to do a proper representative sample. You have to create a kind of miniature of society, and the proportions should be exactly the same as in the whole population. So we, for instance, study the population 18 years of age and older, and we know that 48 percent of them are men and 52 percent are women, we know that on the basis of the statistical census, and in a sample of 1,000 respondents we have exactly 480 men and 520 women. We know 11 percent of people in the Czech Republic live in Prague, so we have exactly 110 people living in Prague, and that applies to all strata of population, age categories, educational groups and so on."

I think it's a source of irritation to many I think when opinion polls are treated as incontrovertible scientific fact. Do you take pains to advise the people who ask you for opinion polls - look this is just a poll, it's not fact, it's not reality?

"Yes, it's an approximation of reality. If one would take it as 100 percent truth, that's nonsense, but this is the best approximation we can do when we want to learn what's happening in a society. I usually say that even an imperfect picture based on solid knowledge is more than sheer ignorance."

Can opinion polls be misused by politicians?

"All spheres of human activity can be misused. It can be well used and it can be misused. No-one would dare to falsify the data of the surveys. But a lot can be done by selecting certain questions, selecting certain answers, and manipulation by selecting specific features and focussing on some aspects and forgetting about others can do most of the manipulation. But this not only applies to politicians. It applies also to journalists and the media. It's well known that if you want to describe especially complicated phenomena in social life, you usually pick up the most visible points and it can be a kind of distortion of complex problems."

You have your finger on the pulse of Czech society - do you enjoy your work?

"Yes, I do. I must say that it's fascinating work. We started early in 1990 with the idea to run a private, independent company so we could take the pulse of the changes in society. The transformation process from a Communist society is unprecedented in history, and from a scientific and professional view it's a fascinating job. A worse aspect is that you study society not only as a scientist but as a citizen. So what's a joy to the scientist is a difficulty for the citizen. But that's life."