Polling In Developing Nations A Challenge

An Acehnese boy plays, as his father watches, in their new home built by an aid organization, Friday, Nov. 30, 2007 in Banda Aceh, Indonesia. Amid the devastation of the 2004 tsunami were some good things, an opinion poll found.
AP Photo/Ed Wray
By Kathy Frankovic, CBS News director of surveys.

Polling is a very American activity-it was invented here and exported along with democracy to much of the rest of the world. We do a lot of polling in the U.S.: at least 24 separate poll reports were released in the four days between the Iowa Caucuses and the New Hampshire primary this year. And despite all the concerns pollsters have about declining response rates and other difficulties, polling in America still seems easy to do. We seem to expect poll results to appear - on any topic - at the snap of a few fingers (plus or minus three percentage points, of course).

This American export is now routinely used elsewhere in the world, both in elections and in non-election circumstances. But pollsters in developing democracies face difficulties unimaginable in the U.S. Though Americans may be overwhelmed by this year's seemingly unending exercise in democracy, we should be aware of the problems other countries face when trying to learn what people think.

I have just returned from a regional conference in India on "Democracy, Disasters and Development" sponsored by the World Association for Public Opinion Research. Tracking the Indian people's relationship with democracy is a difficult process. In the U.S., pollsters know that in many places they need Spanish-speakers among their interviewers. But there are hundreds of languages spoken across India-26 of them spoken by at least 1,000,000 people and 12 of them by 13,000,000 or more. Good pollsters conduct interviews in at least those 12 languages, and newspapers like The Times of India reports on those polls (as well as on polls about the American election).

Research in South Asia has echoes here. The American public had a negative response to the lack of government action after Hurricane Katrina in September 2005; and from that we learned something about the fragility of U.S. society and polity. But Katrina's aftermath also taught us about Americans' willingness to sacrifice to help their countrymen. Fifty-six percent in a CBS News poll conducted in the month after the hurricane said they would be willing to pay more in taxes to help, and the same percentage told the Pew Research Center they had already made a contribution to help those affected. We saw something similar in the aftermath of September 11, when Americans rallied around their country. Survey research also benefited, as (at least for a few months) response rates soared.

The South Asian tsunami of December 2004 similarly exposed strengths and weaknesses in Asian societies. In a paper about that event, Shaan Shankar and Yashwant Deshmukh of Team CVoter compared Aceh, Indonesia, with Sri Lanka, both of which had been devastated by the tsunami. Each place had suffered from ongoing armed conflicts. But post-tsunami, they had different aftermaths.

Aceh had experienced "Hikmah," an Indonesian word that means "wisdom" or "philosophy." One Indonesian expression suggests that in a catastrophe, one can gain something good (it translates as "Do not be too sad; take what happened as a lesson.") And there were good things coming out of the tsunami, according to the opinion poll: not merely the arrival of international assistance, but a greater desire for peace among the public, a feeling of closeness to god and an appreciation for international exposure. This was a "peace dividend," as those on opposite sides of Aceh's longstanding conflict were able to talk with each other. A ceasefire prevailed, and that led to civil discussions.

The informal but very real taxes that had been levied by the warring parties before the tsunami disappeared, freeing up money and income for economic development. Consequently, most people in Aceh today, according to the poll, described themselves as being either as well off as they were before the tsunami or as having the expectation of being better off soon. Thus, even if things were not okay just then, the people of Aceh were hopeful about their future.

But there was no such "peace dividend" in Sri Lanka. Less than a year after the tsunami, armed conflict had begun again, and most people there were not hopeful about their future. Fewer respondents in Sri Lanka than in Aceh felt that things had come back to the way they were before the tsunami, and many people expected things to get even worse.

Polling after disasters is especially difficult. After Hurricane Katrina in the U.S., populations were decimated and scattered. To measure the impact on Katrina survivors, surveys were conducted in Houston shelters, recreating the kind of in-person sampling methodology that was common decades ago here, and is still the only way of polling in many countries. But even so intensive a study could not reach evacuees who had gone elsewhere or had moved in with relatives or friends in other cities.
Polling is difficult enough in multilingual societies, but it's especially hard in places like Aceh and Sri Lanka where armed conflicts have been a fact of life. And yet polling in such places is extremely important for each society's understanding - and may be even more important in the long run for Americans' understanding than slogging through yet another of our pre-primary election polls.

MMVIII, CBS Interactive Inc. All Rights Reserved