Statistics Estonia employed new methodology for this year's population census, using data from several dozen registers to determine a person's physical place of residence. This caused the official populations of the islands to shrink, while Tartu County ended up with more registered residents.
The western Estonian islands lost the most in terms of population as a result of the new methodology, with Hiiumaa now having 9 percent and Saaremaa 5 percent fewer residents. At the same time, Maardu, Loksa and Kohtla-Järve saw incremental increases.
"We looked at whether a mother and father were married, had a joint home loan, whether the mother had gotten a prescription for the father from the pharmacy to demonstrate they are indeed living together. There are other types of connections between people and places. For example, whether people have links to their registered address, power bills, own real estate in the area," Statistics Estonia methodology specialist Helle Visk said.
Personal data is anonymous, meaning that statisticians cannot identity people based on these connections. While people have previously been asked for their place of residence, the data was sourced from registers alone this year.
Population scientists critical of such methods
"Collecting information on people keeps getting more complicated with every passing year. This suggests the data should be put to the test in real life from time to time. /.../ The Nordics are famous for having registers-based surveys and censuses, while they have recently realized that perhaps they should go back to the traditional way of doing things every now and again to check whether the registers are accurate and to what extent," said Mare Ainsaar, senior research fellow of sociology and social policy at the University of Tartu.
She pointed out that Estonians have become increasingly mobile and often own several homes, which is why determining one's place of residence based on registers is complicated. For statisticians, polling people in the traditional way is hugely resource-intensive. However, sample surveys are being conducted to verify the accuracy of methodology the results of which should be in this spring. Of course, none of it would be needed if people noted down their physical address in the population register.
"It is information we need in our practical everyday decisions. Local governments must allocate school and kindergarten places, build roads etc. Knowing who many people we have and who they are is crucial. The benefits directly affect people," Ainsaar explained.
Editor: Marcus Turovski