A basic tool set for a generalized directional model

In political science, there are two classes of spatial models: those which are based on a distance logic and those which are based on a directional logic. This distinction can be found in terms of diverse topics, as voting theory, coalition theory or legislative politics. While Tsebelis (Br. J. Political Sci. 25:289–325, 1995) among others discussed implications of distance models, the respective counterparts for directional models have not been explicitly derived in the existing literature. We try to close this gap by discussing discrepancies between both kinds of models and derive some of the most relevant tools for analyses based on directional models.     

Regierungsabwahl ohne Regierungsneuwahl

Ohne Zusammenfassung Dieser Kommentar zur Bundestagswahl wurde am 4. Oktober 2005 der Redaktion der PVS eingereicht. Die Autoren bedanken sich bei der Redaktion der PVS und bei Thomas Gschwend für Verbesserungsvorschl?ge, die am 12. Oktober 2005 in die Endversion eingearbeitet wurden.     

Sachpolitik oder Parteipolitik?

The article analyzes the roll call voting behavior of German Länder governments in the Bundesrat from 1990 to 2005. We examine if, when, and to what extent the German Bundesrat is dominated by federal party politics rather than by an appropriate conflict of policy interest between states and state governments. We develop and apply a method to separate the effect of policy preferences and of parties’ politics on the voting behavior of Länder governments.     

Hypothesis Testing in the Bayesian Framework

While the Bayesian parameter estimation has gained a wider acknowledgement among political scientists, they seem to have less discussed the Bayesian version of hypothesis testing. This paper introduces two Bayesian approaches to hypothesis testing: one based on estimated posterior distributions and the other based on Bayes factors. By using an example based on a linear regression model, I demonstrate similarities and differences not only between the null‐hypothesis significance tests and Bayesian hypothesis tests, but also those among two different Bayesian approaches, which are also critically discussed.     

Die Eigendynamik zur Eindimensionalität des Parteienwettbewerbs: eine Simulationsstudie

While German ideological space is often characterized with two dimensions, empirical estimation demonstrates that party positions can be placed on a uni-dimensional scale. This seems to be inconsistent with conventional spatial models of party competition, since parties should, accordingly, exhaust the two-dimensional space to maximize their votes. This paper suggests an alternative spatial model which adjusts or removes some restrictive assumptions of the conventional models. Most importantly, the model suggested here allows parties to influence the salience and (non-)separability of both dimensions at voters’ decision processes. The simulation results demonstrate that the vote-maximizing parties can coordinate themselves in favor of inequality of the salience or non-separability of both dimensions. This, in turn, leads to a certain level of unidimensionality of party positions.     

The effect of incumbency on ideological and valence perceptions of parties in multilevel polities

Public Choice - A number of studies recently have investigated party position-taking in multilevel polities. Given the attempts of federally organized parties to tailor their messages to their...     

Personalisierung der Politik in Mehrparteiensystemen am Beispiel deutscher Bundestagswahlen seit 1980

Personalization of politics is investigated as relative importance of leading politicians for the party vote (Zweitstimme) in German Bundestag elections, compared to the importance of parties and the recall of voting behavior in the last federal election. Contrary to recent German research on the impact of special candidate attributes (competence, integrity etc.) we interpret general evaluations of parties and the leading politicians as the most immediate utility attributes or distance measures of the options listed on the ballot. On the basis of this model, of discrete choice analysis as the statistical method and of the Politbarometer data of the Forschungsgruppe Wahlen, Mannheim, we estimate candidate effects in three regions of Germany, characterized by slightly different party systems (West Germany without Bavaria, Bavaria with the CSU instead of the CDU and East Germany with the PDS as a more important party than the FDP or the Greens). It is shown that region is important for personalization, that a general trend towards increasing personalization does not exist, that the most popular politicians sometimes lose their capacity to attract voters to their party and that both attractiveness for floating voters and repulsiveness for former party voters must be taken into account when interpreting effect parameters.