Skip to content
printicon

Historical Population Statistics on Internet

Research project The Swedish population statistics available in Tabellverket for the period 1749-1859 is an essential part of the Swedish heritage. By making this material available on the Internet a valuable resource for research is created.

Head of project

Sören Edvinsson
Professor emeritus
E-mail
Email

Project overview

Project period

2009-01-01 2011-12-31

Funding

Finansår , 2009, 2010, 2011

huvudman: Sören Edvinsson, Demografiska databasen, finansiar: Vetenskapsrådet, y2009: 1071, y2010: 1559, y2011: 1364,

Research subject

History

Project description

Background

The data from Tabellverket 1749-1859, which have been registered by the Demographic Data Base (DDB) at Umeå University, constitute the world’s earliest population statistics. The data consist of two statistical series recorded on preprinted forms representing the size and structure of the population at various time intervals as well as demographic events with particular focus on mortality from year to year. The population forms contain information about population by age, sex, civil status and occupation. The mortality forms show data on the number of deaths by age, sex and civil status. Data are also available on cause of death, births by legitimacy, mother’s age, multiple births, number of entered and dissolved marriages.

DDB has created a research database to allow comparisons over time between the different kinds of information provided in Tabellverket’s forms. From the database, files can be created that are adapted to researchers’ specific questions. For the time being, data are accessible on users’ request only.

On the other hand, each individual form registered from Tabellverket has been made freely available on DBB’s website on the Internet. These forms are searchable by geographic area and point in time. Large amounts of data are thus available and, despite limitations on
accessibility, the material has been used a great deal. For example, during the period May-December 2007, Tabellverket’s website registered 28,600 individual hits. This flow of information quickly becomes unmanageable. Consequently, a user-controlled tool for
handling data and creating files useful for research must be developed. In particular, a method for visualizing Tabellverket’s forms has been requested by researchers and teachers, as visually displayed data simplify exploration of large amounts of information that are otherwise difficult to overview.

Purpose
The purpose of the proposed project is to create an Internet-based tool for accessing the most central demographic data in Tabellverket: total population, births according to legitimacy, deaths, including infant mortality, and migration, all as a function of sex, as well as marriages. This collection of variables, which is referred to as the Base Table, is available for each parish and year. The Base Table will also be extended to contain variables describing the health of
the population. The tool will consist of a cartographic retrieval and visualisation module for creating and exporting files that are useful for research and teaching. The module will serve as a basis for the selection process; however, its most important use will be to visualise selected data as in the atlas work Sveriges Jordbruk vid 1900-talets början [Swedish Agriculture at the Beginning of the 20th Century] (Flach et al. 1909). This is a sought after contribution to the infrastructure that will enable visualisation of complex changes in time and space. It will simplify and greatly increase the opportunities for data analysis and help researchers generate
new hypotheses.

Significance of the project
Considering its size and importance to research in population history, Tabellverket is used relatively little. The sources help us understand demographic changes in pre-industrial society
at a time for which corresponding information is lacking in other countries. It is important to improve accessibility, because a great deal of research on the population in pre-industrial society are still based on sporadic and incomplete estimations made at the beginning of the 1900s by population historian Gustav Sundbärg. The fact that reference is still made to these data is largely due to insufficient accessibility to source material at the parish level.
Digitisation of Tabellverket has significantly increased its accessibility. The proposed project will further improve accessibility by developing a tool for free access to a large number of central variables in Tabellverket. Thus, users will be able to create their own files for research and teaching as well as for visualising results. The cartographic presentation will illustrate
changes in time and space during the period 1749-1859 at an unprecedented level of detail. In this way, it will contribute to new hypotheses concerning the causes of demographic trends and improve our ability to evaluate the relevance of previous research on, e.g., the demographic division into regions during the pre-industrial epoch.
DDB has previously developed teaching materials based on parish registers (DDB, 2006). The present project is also part of our efforts to make historical source material available for
teaching purposes. Access to the Base Table on the Internet will facilitate the work of teachers and students. It will also provide school pupils and university students with opportunities to
explore Swedish population history via Tabellverket’s material.

Researching the source material
The increasing demand for DDB’s long time series data has resulted in, among other things, studies of changed marriage patterns as well as several research projects concerning statistics on cause of death in the form of data on the incidence of certain diseases. Within the Strong Research Environment framework provided by the Swedish Research Council, studies of crises, documented in Tabellverket, such as war, crop-failure and their effects on mortality will increase our understanding of the mortality transition. Other projects within the Strong Research Environment framework will study annual fluctuations in mortality over time as well as spatial processes related to mortality crises.

Data to be made freely available on the Internet will allow studies of population trends in time and space with respect to mortality, including infant mortality, fertility, public health and
migration. To date, very little is known about extramarital fertility, particularly during the 18th century. Completely new knowledge about changes in migration will also be obtained. For the
time being, we lack knowledge about the extent of such changes during the time period covered by Tabellverket, the exception being a small number of individual-based studies from parish data registered by DBB. Public health researchers will have access to an internationally
outstanding study material. The material provides an indispensable demographic foundation that public health researchers can use to analyze social, political and economic changes.
Research questions concerning average length of life in a historical perspective and mortality trends in infants, children or in older population segments will be elucidated through comparisons with modern data. Future research will benefit from internationally unique
opportunities. Nowhere else can epidemiological scenarios be studied at the same level of detail and longitudinality. Researchers with an ambition to develop and clarify preconditions for transitional processes will find in Tabellverket an unparalleled potential for demographic information, which, to date, has been explored relatively little.

Project plan
The project consists of three parts:
1) development of a user-controlled cartographic tool for data retrieval and visualisation,
2) an extension of the Base Table with public health data from
Tabellverket, and
3) creation of user friendly documentation.

1. A cartographic retrieval and visualisation tool for the Base Table
An Internet-based tool will be developed to allow researchers and other users to independently extract, process and present cartographically an optional selection of data (variables, geographic areas and time intervals) from the Base Table.
The user will be able to choose time periods that can be aggregated at an optional level – parishes, counties and other units, as well as the whole country. In addition, it will be possible to select an optional compilation of such areas. The retrieval tool will also allow different kinds of data processing such as sorting, filtering, exclusion or calculation. Use of the tool will be further facilitated by the possibility to save selections for easy access at a later time.
The user-created data files will be compatible with several formats such as MapInfo, ArcGIS, SPSS and Microsoft Excel. It will also be possible to save visualisations as image files for use in later publications.

The cartographic module is created by integrating our tool with a geographic information system (GIS), partly in order to facilitate the selection of data in time and space, partly to illustrate the results cartographically. For example, GIS is used as a tool for presenting geographically linked information. Thus, large amounts of data can be combined and visually presented provided there is a relationship to the geographic areas. Data on geographic and
temporal variations are, thus, presented cartographically when illustration of the data is based on maps. In Sveriges jordbruk vid 1900-talets början (Flach et al. 1909), a cartographic method is used to visualise geographic variations by displaying them as colors or patterns on a map. A similar approach will be used in the present project.

A cartographic display of Tabellverket’s data at a given point in time, or during a certain period of time, provides an
opportunity to observe the material in a new perspective, allowing the user to generate novel hypotheses about societal developments.
The cartographic module will show how parishes have changed over time, i.e., how and when the areal distribution of parishes has changed. For this reason, the module will contain functions for facilitating identification of areas that have remained unchanged during a certain period of time. This information is valuable when selecting data for interpretation of time series related to, e.g., population trends. Mapping the contents of Tabellverket presupposes
that the geographical units in the forms can be linked to spatial objects in existing coordinate datasets. These datasets are, in turn, supposed to contain polygonal data (boundaries) and point data, particularly the geographic parish codes (so-called NAD codes) of the National Archives (Riksarkivet). Because the changes in the parishes and the geographic divisions have already been worked out (RSV - The National Tax Board, 1989), what remains is to create a linkage between the NAD codes and Tabellverket’s parish names.

Development of the cartographic retrieval tool will be carried out in collaboration with HUMlab at Umeå University. It will use as its basis an already developed Internet-based prototype for geographic presentation of historical archive materials. In this way, we can
make use of both available competence and the coordination advantages of using an already developed prototype as our point of departure. The prototype has been developed within the QVIZ project (Query and context based visualisation of time-spatial cultural dynamics),
which is financed by the Sixth EU Framework Programme. The prototype consists of a cartographically based Internet tool for illustrating, searching and publishing archival documents. However, it has not been directly designed to visualise statistics. Further development is therefore needed to make it a tool for illustrating DDB’s unique statistical material found in Tabellverket.

The appearance of the tool will be brought into line with users’ need to carry out their work in a simple manner. A reference group will be formed to guarantee a user-friendly tool.

Work stages:
i. Parish names in Tabellverket will be linked to the corresponding geographic parish codes of the National Archives in collaboration between DDB and the National Archives.
ii. Requirement specifications for the tool will be developed in collaboration between DDB and HUMlab at Umeå University. A reference group of potential users will participate in this work.
iii. Content specifications will be formulated based on the requirement specifications.
iv. Creation of auxiliary tables and reference documents will be carried out in collaboration between HUMlab and DDB.
v. Coding of the tool will be carried out in collaboration between HUMlab and DDB.
vi. The end product will be tested.

2. Extension of the Base Table by adding public health data from Tabellverket
Researchers all over the world, who work with problems related to public health, are very interested in using the unique resource provided by Tabellverket’s statistical material. Adding the pertinent variables will enable studies of Swedish public health both far back in time and in international comparisons. And not least, it will facilitate national comparisons between historical and modern data. Because Tabellverket’s data go as far back as 1749, a number of
public health aspects can be analysed and compared between different geographic areas even in pre-industrial society.

DDB will collaborate with the Swedish National Institute of Public Health (SNIPH) and with Professor Hans Rosling, who is the founder of the Gapminder Foundation and a professor at the Karolinska Institute. Through this collaboration, the Base Table will be complemented
with variables corresponding to researchers’ requests for public health data. Rosling has, among other things, illustrated world health statistics in a very pedagogical way and shown that our image of world health is often based on preconceived notions of developments in different countries.

Variables relevant to descriptions of public health include average length of life, different aspects of child mortality, and age distribution in the productive/non-productive segments of
the population, respectively.

Work stages:
i. In collaboration with researchers at, among other places, SNIPH and the Gapminder Foundation, variables will be selected for use in comparing public health data.
ii. Tabellverket’s base material will be used to create a database with selected variables to be integrated with the Base Table.
iii. The database will be integrated with the already developed retrieval tool.

3. Documentation
i. User-friendly documentation of available variables will be created.
ii. Manuals for the tools will be developed.