Sources & Methodology

In this project, I have used material drawn from 30 archives (including the Berlin Document Center and the Hoover Institution for War, Revolution and Peace) and 35 newspapers. I have compiled the election results from 582 villages and polling places for each party for the 14 elections between 1893 and 1933 (nearly 120,000 independent variables). I have created a second data set with over 200 dependent variables drawn from census data and other sources. When I link these two data sets in a GIS, I am able to map pluralities and percentages of vote received in each election, show areas of party strengths and weaknesses, and display the impact of dependent variables upon the independent variable of percentage of votes per party. It took ca. 1000 hours to compile the two data sets and ca. 1500 hours to create the GIS base map.

Collecting the elections data

    Voting results were ...

Creating  polling place units

    Polling place units in the Oldenburg maintained remarkable consistency throughout the entire period of this study. Polling places in Prussian areas, which showed great variability in the 1870, reached statis in the mid 1880s. In 1912, a small number of new polling places were created, in part to reflect population growth and increased pockets of Social Democratic voters, who the state authorities wished to isolate from farming areas. In 1928, Prussian authorities in the Stade administrative district consolidated smaller places so that Socialist voters in rural districts would be less easily identifiable. I have chosen to use the polling places as the existed in the 1890s as my constant.

Collecting other data

    I have collected the dependent variables from a variety of governmental and non-governmental sources. ...

Coding the data

    One significant problem that I needed to overcome was that the various bits of data available for the dependent and independent variables differed over time. In cases where more than one village were included in a polling place, the location of the polling place could change from one village to the other, sometimes formerly separate polling places were consolidated, or polling places were divided in two. ...