Academic journal article Political Research Quarterly

Measuring Electoral Systems

Academic journal article Political Research Quarterly

Measuring Electoral Systems

Article excerpt

This article compares and assesses four freely available cross-sectional time-series data sets in terms of their information on ballot structure, district structure, and formula of the electoral system in use for lower house and, if relevant, upper house and presidential elections. The authors find that the choice of data source matters for conclusions drawn on the consequences of electoral systems for both party systems and corruption, but that no source can be given prominence over the other on methodological grounds. Students of electoral systems must thus, in the future, make their results sensitive to the choice of data source.

Keywords: electoral systems; data evaluation; party systems; corruption

Few political scientists would deny that the study of electoral systems lies at the heart of our comparative endeavor. A long line of research has established the importance of electoral systems for understanding the workings of democracy (e.g., Duverger 1954; Taagepera and Shugart 1989; Lijphart 1994; Cox 1997; Powell 2000; Norris 2004), and electoral engineering has proven to be a critical issue even under autocracy (Mozaffar and Vengroff 2002; Posusney 2002; Lust-Okar 2006). Yet there is no agreed on authoritative source of cross-national data on electoral systems across the globe. Surprising as it may seem, we lack a single political science data set on a shared set of properties of electoral systems that allows systematic comparisons across countries and over time. The reason for this could hardly be that the necessary theoretical framework for understanding electoral systems is missing. Although the specificities asked for vary depending on one's exact purpose of study, most comparativists seem to agree on some general dimensions along which electoral systems may vary, most notably the ballot structure, the district structure, and the electoral formula (Rae 1967; Blais 1988; Cox 1997; Farrell 2001).

The purpose of this article is to compare and assess four large-n data sets in terms of their information on these three dimensions of the electoral system in use for lower house and, if relevant, upper house and presidential elections. The four data sets to be evaluated are Golder (2005), the Database of Political Institutions (Beck et al. 2001; Keefer 2007), Persson and Tabellini (2003), and Johnson and Wallack (2006). These are the largest and most ambitious timeseries cross-sectional data sets on electoral systems freely available in spreadsheet format.

Two kinds of criteria will guide our assessment. The first we call internal criteria, which pertain to the quality of the descriptive information available. The second criterion we call external, namely, whether it matters for substantial conclusions on external outcome variables what data source on electoral systems we use. We perform this second assessment by exploring the extent to which the different data sources exert different effects in two replication studies of the consequences of electoral systems for party systems and the level of perceived corruption in a country. We find that the choice of data source does matter for conclusions drawn in the external assessment, but that no data source can be given prominence over the other on internal, methodological grounds. Students of electoral systems must thus, in the future, make their results sensitive to the choice of data source.

The article is organized as follows. In the following section we provide a brief overview of the main theoretical dimensions along which electoral systems may vary. We then give an initial presentation of the four data sets to be reviewed, followed by the internal and external assessments, respectively. We conclude by discussing the implications of our findings.

Dimensions of Electoral Systems

There is fairly strong agreement in the literature, going back to Rae's (1967) seminal contribution, that electoral systems may vary along three generic dimensions: the ballot structure, the district structure, and the electoral formula (Blais 1988; Farrell 2001, 6). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.