All researchers involved in this project collaborate in the creation of a new dataset, which gathers data on elections in all administrative tiers of 23 advanced European democracies across a 30-year period (1990-2020). The dataset covers all European countries that have experienced at least 40 years of uninterrupted democratic rule, meaning that it includes both very large countries (Germany, France, Italy) and very small ones (Andorra, Liechtenstein, San Marino). For each country, all governance tiers that are present across the entire country and organize elections are included. While small countries usually have a very limited number of subnational tiers, larger countries have both more tiers and a greater number of units within each tier (the extreme example is France with over 34.000 municipalities). In cases in which the number of units in a given tier exceeds 100, a representative sample (in terms of population size and geography) of units is selected. For each unit included in the dataset, all elections that were held between 1990 and 2020 are coded as separate units of analysis.

               For each election, we include the population and territory size of the unit in which it was held as our independent variables. To measure the unit’s autonomy or degree of decentralization, we rely on scores from the Regional Authority Index (Hooghe et al., 2021) and the Local Autonomy Index (Ladner et al., 2016). We subsequently cover a range of institutional control variables, among which the electoral system, the institution(s) for which the election was held, compulsory voting rules, term lengths, and the executive system. Finally, as our dependent variables, we include measures of participation and competition. For participation we only include voter turnout, which is the only data available across all elections. For competition, we include the number of electoral alternatives, the number of electoral alternatives that were elected (i.e., obtained at least one seat), the number of seats they won, and the percentage of votes they received.

               On the basis of this dataset, we hope to be able to statistically assess the relationship between population size and democratic participation and contestation. The key aim is to discover whether the political patterns we observed during our fieldwork are representative for the total group of national and subnational administrations in Europe. The dataset will therefore play a critical role in enabling us to determine the generalizability of our findings.