The article examines the problem of reliability of statistical data from available sources used in historical and demographic studies.This publication aims to draw the attention of historians and demographers studying the population of Russia in the past to the need for a critical evaluation of sources, including the well-known and comprehensive «Encyclopedic Dictionary» by Brockhaus and Efron (hereinafter referred to as «Encyclopedic Dictionary»).The primary goal of the analysis is to explain discrepancies and misinterpretations of data that arise from the use of statistical information without considering the historical context and results of other studies.The authors compare infant mortality rates in the urban population of Russia during the last decade of the 19th century, calculated from publications of the Central Statistical Committee (CSC) of the Ministry of Internal Affairs of the Russian Empire and according to data from the section «Population» of the article «Russia» in the « Encyclopedic Dictionary». As a result of the study, it was found that when transferring data on the numbers of deaths from the CSC collections to the section «Population» of the «Encyclopedic Dictionary», part of the information was not taken into account or not reflected properly, which caused discrepancies, and the figures were underestimated. Therefore, relying solely on information from the « Encyclopedic Dictionary» may lead to erroneous conclusions that contradict the results of other studies on this topic.Finally, the authors emphasize that when conducting historical and demographic studies, it is essential to refer to primary statistical data, critically analyze secondary and tertiary information resources, rely on the results of previous work by statisticians and demographers, and consider the historical context in their assessments of demographic processes. Broadening the source base through further digitization of historical documents and the use of web technologies will increase the availability and reliability of data.
Troitskaia et al. (Tue,) studied this question.