Enhancing and improving data on party government
The study of party government goes back some considerable time and was for a long time characterized by the use of single case studies or comparing a few, often quite similar cases. Only since the 1970s more systematic comparative analyses emerged and relevant data started to be collected (Dodd 1976). From this time onwards we see more and more data collected (e.g. Budge and Keman 1990; Laver and Schofield 1990; Laver and Budge 1992; Woldendorp et al. 2000). In fact, after 2000 data on party government and related topics rapidly increases. However, as we shall see, the data collected in different places and often for different sets of countries appear, with the help of hindsight, to be of limited use for others or not to fit with other data collections (Döring 2010). Hence, there is urgent need for comparable data and reliable data across the whole of Europe. This chapter will systematically evaluate a selection of widely used data sets on party government in political science research. The main focus is on the diversity and compatibility of the data. The issues that will be addressed are, first, the problematic relation between developing a research question and research design based on specific theoretical considerations that require particular data suited to these considerations on the one hand, and the business of collecting relevant data that would ideally suit a broad array of theoretical approaches, on the other. Second, regarding the context of party government, how to include institutional change (if any) and how to incorporate new theoretical ideas like, for example, veto points or veto players with relevant data in these data sets (e.g. Ganghof 2011). Finally, how to improve the validity (do concepts travel and how far?) and the reliability of
the data collected (are the facts correctly transformed into a data set and indexes?). In particular with the increase of EU members after 1990 this is crucial. Proper and reliable information is not only required for all EU member states, but also in view of the theme of this research volume: Party Government in the ‘new’ Europe. For the evaluation of the data sets, approaches to the study of party government as part of the process of democratic governance (e.g. Budge and Keman 1990; Blondel and Müller-Rommel 1993, 1997, 2001; Keman 2002a; Blondel et al. 2007; Strøm et al. 2003, 2008; Döring 2010) are introduced as point of departure, both to establish the empirical position of party government in that process and to identify the relevant data required for research on its performance.1 The Party Government Data set (further: PGD) is used here as a bench mark but is not considered the best nor the most complete. However, its advantage is that it aims to cover the full ‘life cycle’ of party government. In our view this is an underrated aspect (but see: Strøm et al. 2003; Warwick 2006; Keman 2011a). Too often the focus is only on the ‘birth’ and not at the ‘death’ of party government, let alone on its life course (Mair 2008). Hence the ‘life cycle’ (see Figure 8.1) structures this critical but constructive survey of available data. The structure of the chapter is as follows. In Section 2, the process of democratic governance and the location and role of party government in that process as well as the relevant data regarding party government will be briefly elaborated. In Section 3, the data sets (see note 1) will be located within this process and the data collected evaluated. In Section 4, the data on the institutional context will be elaborated. Finally, Section 5 concludes.