Umeå universitets logga

The primary goal of censuses has always been to collect reliable information on the state’s population and provide a basis for governmental decision-making. This study examines the categories used in the 1930 census and links them to the context in which they were generated. We treat the census as a tool of state power, which can be discerned from the definitions of its categories and the way in which statistics are collected and used. The guiding question of the study was “how does the 1930 census differ from previous censuses and how can these differences and changes be explained?” We find that as in earlier censuses, Statistics Sweden used extracts from the parish books on the individual level to collect information for the 1930 census, but also used diverse supplementary sources including tax registers, income tax returns and language surveys. Thus, unlike in most countries, Sweden did not send out census takers or questionnaires to the population. Many of the new or updated variables we see in the 1930 census such as income, wealth, and number of children born, can be related to the political and social debate concerning the poor working class and the establishment of the welfare state. The inclusion of categories such as ethnicity, religion, and foreign nationality can be seen as part of a normative approach wanting to control, monitor and correct deviant elements of the Swedish population. Sweden has several extraordinary longitudinal population databases built on the country’s excellent parish registers dating back to the 18th century. While the Swedish censuses have rarely been used as sources of data for historical analysis, this work demonstrates that the 1930 census has great potential to support new research.