A hierarchical generalised Bayesian SEM to assess quality of democracy in Europe
- PDF / 1,825,614 Bytes
- 22 Pages / 439.37 x 666.142 pts Page_size
- 40 Downloads / 134 Views
A hierarchical generalised Bayesian SEM to assess quality of democracy in Europe Lara Fontanella1 · Annalina Sarra2 · Simone Di Zio1 · Pasquale Valentini2
Received: 5 December 2014 / Accepted: 29 January 2016 / Published online: 13 February 2016 © Sapienza Università di Roma 2016
Abstract During the last decades, many studies have documented a persistent and widespread decline in satisfaction in some established democracies, giving empirical support to the heterogeneity of democratic systems in terms of their quality. The main implication is that if democracies vary in terms of their performance, such variation should be reflected in citizens’ satisfaction and support for democratic institutions and in citizens’ political trust. In this paper, we examine which factors are related to democracy satisfaction and political trust in the European countries, with a particular focus on the role of quality of democracy in shaping these attitudes. To comply with the aim of the study, we formulate a Hierarchical Generalised Bayesian Structural Equation Model (SEM). The proposed model combines the advantages of multilevel-multidimensional IRT models and SEM and accounts for explanatory variables and indicators at country and individual levels. To explain cross-national variations in the variables of interest, we rely on data from the European Social Survey (EES) and from the Democracy Barometer. The results show that trust in political institutions goes along with satisfaction with democracy. In addition, our findings highlight the existence of a spatial heterogeneous gap between citizens’ expectations and evaluations of democracy across ESS countries.
Electronic supplementary material The online version of this article (doi:10.1007/s40300-016-0081-z) contains supplementary material, which is available to authorized users.
B
Annalina Sarra [email protected] Lara Fontanella [email protected] Simone Di Zio [email protected] Pasquale Valentini [email protected]
1
Department of Legal and Social Sciences, “G.d’Annunzio” Chieti-Pescara, Chieti, Italy
2
Department of Economics-University, “G.d’Annunzio” Chieti-Pescara, Chieti, Italy
123
118
Keywords inference
L. Fontanella et al.
Quality of democracy · Multidimensional IRT models · SEM · Bayesian
1 Introduction Over recent years many social science scholars have devoted their efforts to measure citizens’ attitudes towards democracy and governance. The correct understanding of citizens’ support and satisfaction with democracy unavoidably implies knowing what democracy is. In literature, there is a general agreement that this is a broad notion, difficult to conceptualise and, even if the bulk of the empirical research considers a unidimensional construct (see,forinstance, [17]), some recent studies adopt a multidimensional approach [7], deeming democracy as a multi-faced concept, not well characterised by a single indicator. A greater nuance and precision of its connotative meaning can be achieved by looking at the emerging literature on the “quality of democracy”. In th
Data Loading...
