Statistical data of the Weimar Reichstag 1928-1932 elections results: on the interpretation of the sources

The article is devoted to the Reichstag elections in the Weimar Republic in 1928-1932 and the problem of sources of statistical data, discrepancies and inaccuracies of different interpretations of these data. The author reveals four groups of data and their interpretations, inaccuracies in the data researchers used, and gives some new calculations and conclusions. In particular, the turnout of voters in percentage, the number of non-comers in the sum with invalid bulletins, the percentage data on vote for the main parties of the Weimar Republic were revised. The author divides the main existing sources of data on the results of the elections to the German Reichstag of 1928-1932 into four groups, depending on the proximity of their data and techniques of calculation. The first group includes the Soviet historical encyclopedia and three German editions: Dokumente zur Deutschen Geschichte 1929-1932, Deutsche Geschichte 1933-1945 and Dokumente zur deutschen Verfassunggsgeschichte. With the general basic similarity in this group, there are discrepancies, but they are not so considerable. The second group is represented by a study of Horst Moeller, in which the results of vote on the parties and the voter turnout in percentage terms differs essentially in the majority of positions from the other groups. The contemporary German electronic edition at gonschior.de/weimar/Deutschland, in which data generally correspond to the first group, yet have some essential differences, represents the third group. The fourth group includes Wahlen und Abstimmungen in der Weimarer Republik: Materialien zum Wahlverhalten, 1919-1933 by J. Falter, T. Lindenberger and S. Schumann. The author comes to a conclusion that the percent of the voter turnout is more exact in groups 1, 3 and 4 and is incorrect in Statistisches Jahrbuch fur das Deutsche Reich and in Moeller's study. Data on the general voter turnout in percentage terms are most exact in the first, third and fourth groups. They (in the first group with the corrections and specifications given in the article) give correct percent of votes for the parties. Data on bulletins recognized valid and invalid are truly reflected in the third and fourth groups. The author makes calculations according to which the number of those who did not come to polling precincts for vote or whose bulletins were nullified differ from calculations made by Heinrich Shtrifler and accepted in science. The difference between the elections of 1928 and 1930 was 2.5 million votes, but not 2.9 million votes, the difference between the elections of 1930 and July, 1932 was 0.6 million instead of 0.9.million. The difference between the valid bulletins in the July and November elections of 1932 was 1.56 million, but not 1.6 million, as Schtrifler considered.

Keywords

Authors

References