Measuring the change (or the lack of) of political attitudes 1 |
|
| Chair | Dr Roula Nezi (GESIS-Leibniz-Institute for the Social Sciences ) |
| Coordinator 1 | Dr Theofanis Exadaktylos (University of Surrey) |
Ethnic parties are often viewed with concern and have been banned in several countries for fear of polarisation. Some argue, however, that ethnic parties may be valuable vehicles in solving ethnic tensions because they contribute to the integration of diverse ethnic groups into the national political community. Yet few studies directly examine ethnic parties’ contribution to societal polarisation or integration and, where they do, cross-sectional data do not allow inferences with regard to the direction of causality. This paper examines original longitudinal data from a fixed panel survey surrounding the Romanian elections in December 2016. Romanian elections provide a fruitful context to analyse the effect of ethnic party campaigning on voters’ attitudes due to the presence of established Hungarian minority parties. The panel data captures the attitudes of representative samples of voters with different ethnic backgrounds both before and after the election campaign: i) Romanians in the majoritarian Hungarian provinces of Harghita and Covasna; ii) Romanians in the rest of the country; and iii) Hungarians in both Harghita and Covasna and neighbouring provinces. Within-group, over-time comparisons can establish whether the election campaign has led to changes in political attitudes. Between-group comparisons can show whether the status of the each group as local majority or minority mitigates the size or direction of change. To adjudicate between the polarisation and integration hypotheses, the paper uses multilevel modelling techniques to consider individual and contextual characteristics as well as changes over time in a single model. It examines differences within and between groups with regard to four key dependent variables covering ethnic relations (belonging to one’s own ethnic group and evaluation of the respective out-group) as well as political attitudes (belonging to the national political community and support for the political system). If the polarisation hypothesis is correct, the heightened salience of ethnicity during the elections should increase Hungarians’ belonging to their in-group, worsen evaluations of the Romanian out-group, and decrease belonging to the national community as well as support for the political system. Romanians in the majoritarian Hungarian provinces should feel threatened, affecting their evaluations of the Hungarian out-group. If the integration hypothesis is correct, ethnic relations and political attitudes should remain stable or even improve, particularly among the Hungarian minority. As data collection is under way at the time of writing, no preliminary results exist yet.
From the latest waves of World Value Survey (WVS 6) and European Value Survey (EVS 4); the study tried to derive the trust level of the respective countries’ government. Here, trust in government means combined confidence level on three main organs of a government: executive, legislative and judiciary. According to the performance based argument and/or quality of governance argument; OECD countries are supposed to have higher institutional trust compare to other countries. However, the survey responses showed that there is a good number of countries where the government have higher level trust despite poor performance (measure through Human Development Index, 2015 by UNDP) and poor governance (measure through Corruption Perception Index, 2015 by Transparency International). The current study likes to explore the possible explanation of these dynamics based on the variation of individual’s perceived notion on state: right seeking vs privilege seeking. Here, ‘right seeking’ means citizens who tend to believe that access and availability of different public services as their right while ‘privilege seeking’ citizens tend to believe that getting different services from different public institutions is a privilege and act of generosity of the state. Based on this distinction on individual’s perception, the study found statistical significant differences on the level of trust in different institutions (like president/prime minister, parliament, and police) in 34 African countries from 51,587 respondents. The study also tries to explore further about the possible mechanisms of such trends. The study found that people with ‘privilege seeking’ orientation tend to believe that there are less governance related problem than ‘right seeking’ people (like less corruption or unequal treatment) while people with ‘privilege seeking’ orientation have higher rating of different government performances than ‘right seeking’ people (like better management of economy and reduction of crime). The study also found that ‘right seeking’ orientation want to contribute more for state (through taxation) than ‘privilege seeking’ people.
In this paper, we investigate how natives’ acceptance of immigrants in Germany changed (1) with certain characteristics of immigrants and (2) before and after New Year’s Eve 2015/16 as well as (3) the interaction of both aspects. On that night, several dozen incidents of sexual assault and robbery took place at the festivities around the central station of Cologne as well as in other German cities. These assaults were clearly related to the inflow of refugees originating from Arabic and North-African countries by many public speakers and the media. Drawing upon group threat-theory and multi-level agenda setting-theory, we argue that it is important to pay attention to the central arguments constituting the public debates following such events because they affect subgroups of immigrants differently. We conducted a two-wave factorial survey experiment where respondents were asked to rate a set of hypothetical immigrant profiles consisting of six attributes. The first wave of our survey took place in Spring 2015. To investigated potential changes in the effects of the immigrant attributes, we replicated this survey in February 2016 using the same sample of respondents, and thus drawing upon intra-individual variation for our analyses. First, we assure that our dropout is not selective with respect to relevant covariates (including the average vignette rating in wave one). Subsequently, we employ three-level hierarchical linear models to estimate the main effects of the immigrant attributes as well as their interaction effects with the event treatment dummy variable. In total, our final sample consists of 719 respondents rating 14 immigrant profiles at two time points, leaving us with 20,132 unique ratings. Generally, the respondents rated immigrants 0.26 points more negative on a 7-point scale, which amounts roughly the negative effect Muslim immigrants faced as compared to non-religious immigrants. Moreover, immigrants of African and Arabic origin were rated significantly more negative after the assaults and the initial preference for male immigrants diminished. Interestingly, deeper analysis reveal that the latter is mainly true for female respondents. The negative effect for Muslim immigrants, however, is not statistically significantly stronger after the assaults and, strikingly, migrants escaping political persecution (“refugees”) were accepted even more at time point two. The effects of aspect that were not related to the public debates, such as language skills or qualification, did not significantly differ before and after the assaults. We conclude that individual attitudes respond to public discourses and to the framing of certain events. Methodologically, vignette studies are ideally suited to disentangle the different aspects of such debates in order to investigate their impact on public opinion. Substantively, the individual rejection of certain immigrant subgroups on the one hand corresponds to the general national discourses of the topic. On the other hand, humanitarian aspects seem to be deeply internalized so that the support for persecuted refugees is even stronger after the assaults.