The analysis of regional variations of voting behavior (turnout and party votes) requires territorially disaggregated data, namely, election results at the level of single constituencies. Data used for analysis are those published in machine-readable form in the CD-ROM EWE-1815 and documented in the accompanying handbook. As the present book analyzes for the first time that wealth of new and unexplored data, the following section gives a summarized description of the collection and of the criteria that guided it. The rest of the chapter is concerned with methodological aspects, indicators of the formation of national electorates and party systems, and measures.

Data

Periods Covered and Data Sources

The period of time covered by the analysis is approximately 150 years, roughly from the democratic revolutions of 1848 – a crucial step toward parliamentary democracy in most West European countries – to the present. However, among the 17 countries considered, the period of time varies according to (1) patterns of state formation and (2) availability of sources (see Table I.1 in the Introduction and Table 2.1). The period of time covers all elections ending with the most recent elections published by 1999 (and does not include nondemocratic elections of the Nazi and Fascist periods between the two world wars).

As far as patterns of state formation are concerned, the starting point is determined by the timing of national unification or independence and by the definite transition from estate (or absolutist) systems to modern parliamentary systems based on general territorial representation.