By Mary-Russell Roberson
After the 2004 Indian Ocean tsunami, Elizabeth Frankenberg and Duncan Thomas collaborated with their colleagues at SurveyMETER, an NGO in Indonesia, to develop a longitudinal survey in Aceh to study the impacts of the tsunami and to track recovery. In developing the project, they worked with a government institution called Statistics Indonesia, which provided access to household data collected shortly before the tsunami.
The resulting project, called the Study of the Tsunami Aftermath and Recovery (STAR), has been following 30,000 survivors ever since. The team conducted surveys every year for the first five years and is currently conducting a 10-year follow up.
Out of that group of survivors, 96% have participated in at least one follow-up survey, despite the fact that many lost their homes in the tsunami and have moved more than once since then.
PhD student Nicholas Ingwersen visited Aceh with Frankenberg in 2005.
Highlight for Basic Page
A child receives a STAR project health check in spring 2014. The researchers follow 30,000 survivors of the tsunami. “The survey is raising lots of questions to which we do not have answers,” said Professor Duncan Thomas, “But we will after 10 or maybe 20 years. Our goal is to follow them for 100 years, forever.”
“We went to this one place where we were having trouble finding anybody,” he said. “On the map, what was a road is now the sea; where the house was is now a beach. It was an extraordinary effort to try to find these people.”
In fact, that first survey took almost a year. But Frankenberg and Thomas were persistent because they believed that the people who were hardest to find included those who had been most affected. “We put our resources into turning over every last stone,” Frankenberg said.
In the years since, the team has continued to refine the survey and the process. Local interviewers—250 of them—are trained to interview with precision, patience, and empathy, spending 4-5 hours with each household. The interview data is entered into a secure computer system, which includes features to help keep track of respondents. For example, if an interviewer discovers a respondent has moved, the new address is automatically shared with the interviewer of the appropriate territory.
The dataset allows comparisons between people from communities that were affected to different degrees. The team used satellite images and on-the-ground interviews to categorize each community as heavily affected, somewhat affected, or not directly affected.
The robustness and sheer volume of the survey data create opportunities for research into a wide variety of topics. Ingwersen’s PhD work illustrates the range: he’s doing one project on economic choices among adult survivors, and another on the growth of children who were in utero during the tsunami.
“It’s neat to be able to work on a project where you can do all these different things with the data, especially from the point of view of policy,” he said. “There are all these different ways these people’s lives can be affected, and they’re all really important.”
Frankenberg and Thomas plan to make all the data publicly available to the scientific and policy communities. Read more about their research on the STAR project.