Seminar "On the reliability and comparability of the results of the 2021 Russian Census"
On June 8, PhD, Head of the International Laboratory for Population and Health, Evgeny М. Andreev at the seminar "Modern Demography" spoke about the reliability and comparability of the results of the Russian population census of 2021.
On June 8, PhD, Head of the International Laboratory for Population and Health, Evgeny М. Andreev at the seminar "Modern Demography" spoke about the reliability and comparability of the results of the Russian population census of 2021.
At the beginning of the seminar, it was noted that the All-Russian population census 2021 stands out among other Russian censuses in that it took place during the period of the highest incidence of COVID-19 in the country, which raises concerns about the incompleteness of population registration in the census.
The report compared the reliability of the results of the 2002, 2010 and 2021 censuses. Starting from the 2010 census, it is officially possible to conduct a census according to administrative sources (only gender and age were allowed to use), which led to an increase in the proportion of persons with unknown education and marital status. The share of such persons in the results of the 2021 census was approximately 16% compared to 3% in 2010.
Another important question was about the comparability of the results of the 2021 census with the 2010 census data. Thus, connection was found between the proportions of those according to administrative data and the discrepancy between the census results and the estimated population of children, which suggests that the discrepancy is a consequence of incomplete registration of children in administrative databases. It seems that the most correct thing would be to focus on the results of the estimated population at the age of up to 15 years.
But it is most difficult with the elderly. It has been repeatedly suggested that the 2010 census overestimated the number of elderly. It is most likely that the 2021 census confirmed this assumption. Administrative sources were probably clarified. But the revision of the results of the 2010 census is hardly possible. It is more rational to recalculate the estimates for 2011-2020 and accept the non-smoothness of the series between 2010 and 2011.
E.M. Andreev believes that statements about the unreliability of the 2021 census results are not justified by anything. The results of the census are not perfect, but in general, the accuracy of the 2021 census is not lower than the 2002 and 2010 censuses.
Bulat Shamsutdinov, research assistant.