Show description

The systematic sample organised and supervised by Brustein and Falter, the second largest 36 The Social Bases of Nazism, 1919–1933 of the BDC-based samples involving 42,004 cases [20: 17; 21: 88], is probably the nearest to a representative sample extracted from the BDC holdings to date. Reservations have also been expressed about the representative nature of the results obtained by different scholars who have dealt with the question of the social geometry of the Nazi electorate, as well as the methodologies employed by them [94: 165–94; 82].

The value of the first representative analysis of the Nazi electorate by Courtney Brown [19], who examined all of the 946 counties of the Weimar Republic, is limited by the fact that he dealt only with the July 1932 Reichstag election, used too few variables and excluded the large electoral group ‘assisting family members’ from his calculations [94: 180–1]. The utility of the collaborative analysis by O’Loughlin, Flint and Anselin is similarly limited Methodological problems 37 by being restricted to the Reichstag election of September 1930 [110].

Andrews’ suggestion of a classification model divided into seven social categories and thirty-eight occupational sub-groups does, however, raise practical difficulties when it comes to publishing the huge tables which would be generated by such a model. There is little to suggest that the historians and political scientists who produced the first quantitative studies on the sociology of the Nazi Movement in the 1970s were guided by the massive amount of information on the social structure of the working population of Weimar Germany which is to be found in the introductory volumes to the census returns of 1925 and 1933.