Sisp Conference 2019

Section 11 - Metodologia della ricerca (Research Methodology)

Managers: Luigi Curini (luigi.curini@unimi.it), Vincenzo Memoli (memoli@unict.it)

Thursday 12th September 2019
  Donato Valli - Aula B 10.15-12.00
  Monastero - Padiglione Chirico 10.15-12.00
  Donato Valli - Aula 6 13.15-15.00
Friday 13th September 2019
  Monastero - Sala gradonata 14.00-15.45

Chairs: Edoardo Guaschino, Alessia Damonte, Renata Lizzi, Maria Stella Righettini

Discussants: Alessia Damonte, Renata Lizzi, Nicola Righetti

Fuzzy-Set Qualitative Comparative Analysis of Russia’s Foreign Policy in the Near Abroad: a Research Design
Adriana Cuppuleri
AbstractSince the fall of Soviet Union, Russia has adopted various instruments in order to maintain a regional primacy in the near abroad: some of them have been more assertive (e.g. Crimea and Georgia) while others have been more cooperative (e.g. Belarus and Armenia). How to explain this variation? Scholars usually map Russia’s foreign policy according to several theoretical models, which are respectively generated from the individual, the state and the structural levels of analysis in International Relations. To account for the complex interplay between causal factors that include external interference (I), status recognition (S) and state capacity (C), this article develops a neoclassical realist framework of great powers’ neighbourhood policy. This paper applies fuzzy-set Qualitative Comparative Analysis (QCA) on original data about 18 Russia’s policy responses in the near abroad between 1992 and 2016. The findings report the existence of multiple path toward assertiveness: they show that external interference from competing great powers plays a large role compared to domestic and ideational factors. At the same time, external interference leads to assertiveness only when combined with social status and state capacity.

The Economic Crisis, Austerity and Public Health Systems in Europe: Assessing the Impact on the Right to Health through fuzzy set Qualitative Comparative Analysis (fsQCA).
Rossella De Falco
AbstractThe 2008's global financial crisis has undermined public health systems in many European countries. In Greece, Italy, Spain and the UK, harsh austerity measures included the containment of health spending, resulting in higher out-of-pocket payments (OOPs), increased waiting lists as well as shortage of medical staff and facilities. This, potentially, increased barrier in accessing healthcare, with vulnerable groups bearing the highest cost of the crisis. There is much research on the effects of the economic crisis in single countries, but comparative empirical research is much more limited. Therefore, this paper attempts to verify whether the countries that decreased public health spending also present higher unmet medical needs through fuzzy set Qualitative Comparative Analysis (fsQCA). Specifically, the 27 European countries will be ranked by their degree of membership into 'health austerity' sets, through the analysis of both public and private health spending. Then, it will be seen if there is an association between the group of countries that reduced public spending during the crisis years and countries that presented an increase in unmet medical needs due to waiting lists, financial reasons or distance. This latter indicator is collected by the EUROSTAT through the EU-SILC yearly survey. As there might be multiple paths leading to higher health barriers, QCA is particularly appropriate for this field of research. In fact, QCA allows for complex causality and equifinality in both small and large-N research desings. Access to healthcare is a fundamental human rights indicator when it comes to the enjoyment of the highest attainable standard of physical and mental health. Therefore, the aim of this paper is to assess the impact of post-crisis economic recovery policies on one of the tenets of the right to health through fsQCA. Set-theoretic empirical research is then combined with process-tracing of some cases, as to highlight the causal link at national level.

Against all odds: the curious case of EU’s food aid policy in historical perspective.
Ilaria Madama
AbstractBackground - The paper deals with European food aid policy, which has now become a highly symbolic component of the “social dimension” of the European Union. More in depth, the paper offers a reconstruction in historical perspective of the evolution of EU’s food aid policy, from its origins in the late 1980s to most recent transformations, highlighting elements of continuity and change. Firstly introduced in 1987 under the Delors’ presidency of the European Commission, the European Food aid program for the most deprived persons (MDP) was meant to serve as a way to handle – meeting at the same time social and societal ends - the surpluses generated by the Common Agricultural Policy. Over time, however, the program incurred in a series of reforms that altered its functioning and, partly, its scope. This is particularly evident with the launch of the current program, the Fund for European Aid to the Most Deprived (FEAD), which in 2014 marked a major discontinuity. Aims - The paper has a twofold aim. First, it offers an in depth analytically-driven illustration of the institutional path of EU’s food aid policy, which displays elements of continuity but also gradual adaptation to changing opportunities and constraints. Second, the study aims at investigating the political and institutional dynamics that have characterized this policy sector, shaping its changes at three key moments (1994, 2008, 2014). Methods - The article deploys a qualitative approach. Historical analysis, through process tracing technique (relying on documentary analysis plus a number of semi-structured interviews), was used to identify key moments in the sequence of institutional change and to examine the decision making process, allowing to shed light on specific causal mechanisms at work. Findings - Preliminary findings suggest that EU's food aid policy, despite its resiliency, resulted to be a contested and contentious program, that fostered the emergence of harsh vertical and horizontal tensions. Nonetheless, the institutional and political sponsorship proved to be strong enough to have the scheme not only maintained but even strengthened, in terms of scope and financial budget, over time. Due to these features, EU’s food aid policy is a puzzling case, which lends itself to represent a key case-study to unveil complex political struggles arising around the making of «Social Europe».

Xylella politics: How everything went wrong and what can we learn
Simone Busetti
AbstractThe paper aims at reconstructing the decision-making and implementation of crisis management related to the emergency of the bacterium Xylella Fastidiosa in the Apulia region. Xylella is deemed responsible of the so-called Olive Quick Decline Syndrome (OQDS), a disorder of olive trees that appeared in 2013 in a limited area in the province of Lecce in South-East Italy and has not stopped expanding until today. Since its recognition in Apulia, strains of Xylella different from the one in Apulia have been discovered in Switzerland, France, Spain, and Germany. The Xylella emergency is an interesting case for several reasons. First, the science base is contested. The disorder appears to be complex, arising out of the simultaneous presence and interaction of three parasites (a lepidopteran, a set of fungi, and the bacterium Xylella) and preferring some species of olives with respect to others. The complete novelty of the disease and the limited evidence available did not help responding to the crisis. Stakeholders, media, political parties and institutions even doubted that Xylella was truly responsible for the disorder and some still doubt it. The ground was fertile for conspiracy theories, exemplified by a judicial investigation accusing researchers to be the true responsible of the disease. Second, olives in the province of Lecce are not common trees. They represent a fundamental asset of the economy of that territory and they are protected as part of the historical landscape of the area. The economic and environmental values of olives set the stage for a fierce opposition by business, local stakeholders, and environmentalists, all committed to avoid the eradication of infected trees. Uncertain knowledge-base, failure in communicating risk, conspiracy theories, opposing interests and the combination of social, economic and environmental effects produced a dangerous mix for the public authorities responsible of the emergency. This resulted in several delays in setting up containment measures, obstacles during implementation and conflicts in the multi-level governance of risk. The paper will reconstruct the decision-making and implementation of the emergency in order to investigate the role of different factors in influencing public response and draw lessons for future crises.

Chairs: Mauro Barisione

Discussants: Cristiano Vezzoni

Almost terror: Macerata shooting and the stillbirth of a Digital Movement of Opinion
Guido Anselmi
AbstractOn the third of February 2018, far-right sympathizer Luca Traini opened fire on a group of black migrants in the Italian town of Macerata, wounding six persons. The violent act was motivated by the need to ‘avenge’ Pamela Mastropietro, a 19 years old italian woman brutally killed, two days earlier, by her black migrant pusher. While the incident resulted in no deaths, due to the superheated political climate (national elections were due in a month) the aftermath of the shooting immediately generated a huge social media controversy, in which the contested issue was the labelling of the incident as an act of xenophobic terrorism. Following an early claim to action by writer and Twitter influencer Roberto Saviano, anti-xenophobes clashed (mainly over Twitter) with right-wing sympathizers over the definition of the shooting: the formers framing the event as a consequence of rising xenophobia, while the latter framed the shooting event as an extreme consequence of diffused insecurity, caused by the permissive migration policy choices brought to be by the left. While the earliest days of the flame (3rd and 4th of February) are marked by a success of the first frame; with the passing of time, the framing of the shooting becomes highly polarized, contested and divisive. The main research interrogative of this paper deals with the conditions in which a Digital Movement of Opinion (see Barisione et al 2017) may come into existence. The main hypothesis is that the lack of an external shock (i.e. actual deaths), in combination with the pre-existing polarization, contributed to the collapse of the potential DMO: as circumstances did not generate strong enough affective investment in the anti-xenophobe frame while, at the same time, not silencing contrasting positions. Empirical data comes from 571996 italian language tweets, gathered from 3-2-2018 up to 28-2-2018. To operativize framing by Twitter users a machine learning approach has been deployed. In order to assess the internal consistency of crowds in both frames fragmentation of the retweet network has been calculated, this allows for a modicum of triangulation and hence more robust conclusions. To further triangulate empirical findings a quick comparison with a smaller set of italian language tweets mentioning ‘Luca Traini’ during the aftermath of the Christchurch mosque shooting will be provided.

How short is short? Assessing the (non)-persistent effect of four terrorist attacks on attitudes toward immigrants (2015-2018)
Moreno Mancosu, Mònica Ferrìn Pereira
AbstractGrowing research focusing on citizens’ psychological reactions to terrorism finds that attacks perpetrated by individuals belonging to Muslim minorities increase negative attitudes towards those perceived as the “outgroup”. However, few attempts have been made so far to assess the temporality of the impact of these attacks. While we know that immediate emotional reactions are deemed to cause stereotyping effects in the days following traumatic events, we need more research to assess how long this effect endures. By means of a quasi-experimental before-after design based on four terrorist attacks (the Bataclan attack of November 13, 2015, the Berlin truck attack of December 19, 2016, the Manchester bombing of May 22, 2017, and the Carcassonne and Trèbes attacks of March 23, 2018), we ask whether the stereotyping effect occurs, and how log does it takes (if ever) to diminish its strength. We present evidence consistent with the stereotyping effect hypothesis, but we also show the role of temporality: while the effect is strong and significant in the first few days after the attacks, it disappears after about 72 hours.

To perceive and to believe: misperceptions about immigration and attitudes toward the EU
Francesco Visconti
AbstractPublic opinion towards immigration has been increasingly scrutinized by scholars because it embodies one of the major divides within contemporary western societies. It has become central in the political debate on European integration and has been cast as one of the three main ‘crises’ of the EU, together with the Eurozone crisis and Brexit. This is not only a consequence of the 2015 refugee crisis, but also of the constant increase in intra-EU mobility and migration from non-EU countries that took place in the last few decades. The literature argued that individuals who feel themselves more economically and culturally threatened by the growing international mobility tend to express both more Eurosceptic and pro-closure preferences. This article investigates a complementary explanatory factor which relates to citizens’ information about immigration. In particular, it looks at how the misperception of the actual number of immigrants has changed across time in European countries. Moreover, it looks at its association with citizens’ attitudes towards immigration and their support for the European Union. It does so using both a cross-country and longitudinal perspective employing survey data from Eurobarometer and European Social Survey.

Do debates about the immigrants affect differently men and women? An assessment of immigration-related media claims on female attitudes to immigration.
Lenka Drazanova, Leila Hadj-Abdou
AbstractAcross Europe, debates about migration have become ever more salient in recent years. Among other issues, immigrants have often become associated with practices of female oppression and disrespect for gender equality. These fierce debates have consequently led not only to policy reforms in the domain of immigrant integration, but have also legitimized demands for restrictive immigration control in numerous European countries. In this study, we explore whether debates regarding immigration have an impact on attitudes of the group which has been portrayed as most ‘vulnerable’ and threatened by immigration: women. We ask whether the intensified association of immigration with gender inequality in public discourses plays a role in how women view immigration. We examine the issue by looking at the development of female attitudes to immigration over time, particularly focusing on the link between these attitudes and the frequency and direction (positive-negative) of immigration-related news reports. We use data provided by the eight rounds of the European Social Survey (2002 – 2016) together with country-level information regarding the media environment from the ESS Media Claims data. Utilizing multilevel modelling in 15 European countries over time, the paper accounts for cross-national differences in women’s attitudes to immigration and especially cross-national differences in the relationship between gender and the frequency, as well as type, of immigration-related news reports. Our findings indicate that the frequency of immigration news in the media (regardless whether the content is negative or positive) is associated with majority women’ negative attitudes towards immigration. In addition, our findings indicate that in countries where respondents are exposed to media framing immigration as beneficial to the society, women become significantly more positive to immigration compared to men. Overall, these findings bring us one important step closer towards a better understanding of the gendered aspects of inter-ethnic relations between majority members and immigrants in European host societies and the role media might play in these relationships.

Chairs: Nicolò Conti, Andrea Pedrazzani

Discussants: Davide Angelucci

Policy congruence within parties: Bound to be divided?
Andrea Pedrazzani, Nicolò Conti
AbstractHow unified parties are on policy and the ways in which distinct party organisational faces contest different issues are key questions for party politics research, especially if one considers the phenomenon of increased personalisation of parties that may introduce linkage disequilibria among different party faces. Whereas empirical distinctions between various types of party actors with specific concerns at the political level have gone relatively unnoticed in the literature and different party organisational components have been implicitly treated as identical actors, thus assuming the same strategic calculations for all types, we show a mounting tension inside political parties between a leadership that is more sensitive to popular pressures and public office holders that do not necessarily share the same views of the leadership. In the paper, we analyse policy congruence between party leadership and party elites serving in public office in Italy. For this purpose, we make use of a triangulation of data based on the Chapel Hill Expert Survey and the Comparative Manifesto Project estimating party positions, and on original data from an elite survey conducted in 2019 estimating the positioning of Italian parliamentarians.

Quantifying dynamic policy responsiveness: Evidence from state minimum wage laws in the U.S.
Gabor Simonovits
AbstractTo what extent are public policies responsive to changes in mass preferences? While existing studies provide evidence of an association between changes in preferences and policies, they offer an incomplete account. Because research on dynamic representation measures preference change and policy change on incomparable scale, it fail to distinguish between under-responsiveness (policy change is to slow compared to opinion change) and hyper-responsiveness (policy change is overly fast). This is problematic as depending on the baseline divergence between preferences and policy outcomes, the normative appeal of under- or hyper-responsiveness is ambiguous. In this study we overcome this problem by estimating average preferences about the minimum wage in each of the 50 U.S. states in 2013, 2016 and 2016. In particular, we combine the use of open ended survey questions - to elicit preferences on a dollar scale - and using dynamic multi level modeling with post-stratification - to obtain state by year measures of preferences. Matching the resulting panel dataset with corresponding data on minimum wage laws we set out to quantify the extent of dynamic responsiveness and characterize the evolution of policy bias across the states.

Votes, Ideas, and Linkages: Mass-Elite Ideological Congruence in Brazil
Joaquim Meira
AbstractHow congruent are the opinions of Brazilian citizens and representatives, and how does this congruence vary across themes and groups? The principle that public policy and government decisions must respond to the wishes of the population is typically pointed as key to democratic regimes. However, there are rising claims of a “crisis of representation” in Brazil and Latin America in the last decades, with citizens losing faith in being represented by government and politicians. Plus, although the correspondence between citizens’ wishes and representatives’ thoughts and actions is a major topic in political science, studies on the Brazilian case are scarce. The goal of this study is to map the ideological congruence in Brazil during the 54th Legislature’s timespan (2011-2014), answering the following questions: (1) in which themes there is more congruence, (2) whether the country’s main parties work as channels of representation of interests, and (3) if the preferences of citizens of higher socioeconomic class are better represented in Congress. In order to do that, I employ novel methods and data with the potential to yield robust and reliable results. The data sources are national sample and legislative surveys (the WVS and the BLS, respectively). The main method is the Earth-Mover’s Distance, which measures the degree of similarity between data distributions, and presents better accuracy and less information loss than most alternatives employed by the literature. I understand congruence thus as the degree with which the distribution of preferences inside the legislature resembles that of the population. My results are that congruence varies a depending on the thematic dimensions, with traditional economic issues being very congruent, ``new issues'' not far behind, and ideological self-placement and ``libertarian'' ideology being less congruent. I also find that neither of the two main parties promote congruence between their partisans and elites, and that Congress is more responsive to citizens with a higher educational level. My findings contribute to the understanding of the Brazilian representative system, to comprehending which interests get voiced in Congress, and to the methodological debate on ideological congruence.

Who gets represented? Patterns of MP-voter congruence on European Solidarity in the core and periphery
Alessandro Pellegata, Francesco Visconti
AbstractIn the aftermath of the Eurozone and the refugee crises the EU witnessed a growing tension between the social and economic dimensions of integration. Against this background, this article investigates the congruence between voters and incumbent members of national parliaments on the highly contentious aspects of EU solidarity. First, we map the extent to which national political elites share similar views with their voters by looking at cross-country differences in MPs-voters congruence. Secondly, we assess which factors contribute to explain levels of MPs-voters congruence testing individual, party, and country level factors. Employing original data from elite and mass surveys conducted in six EU countries by the REScEU project, we find that both voters and MPs express high level of support for policies strengthening European solidarity in the Eurozone periphery. Instead, citizens of core member states share more positive preferences than their national representatives. Furthermore, this gap is higher between radical-right Eurosceptic MPs and their voters than between voters and MPs of other parties

Chairs: Cristiano Vezzoni

Discussants: Riccardo Ladini, Moreno Mancosu, Federico Vegetti

A gender gap in misperceptions? Comparative evidence from cross-national surveys
Riccardo Ladini, Moreno Mancosu
AbstractOver the last few years, the spreading of misinformation and fake news has increased scholars’ attention toward the accuracy of people perceptions about societal issues. The diffusion of false information is indeed expected to be easier when individuals have inaccurate perceptions of the world that surrounds them. Several comparative survey projects measured misperceptions of societal facts that can be assessed objectively, by asking individuals to guess the size of the immigrant population in their country (European Social Survey 2002, 2014; Eurobarometer 2017) or the unemployment rate (European Social Survey 2008) at the time of the interview. Previous research has investigated the impact of several individual and contextual factors in predicting misperceptions, but little attention has been devoted to the role of gender, generally included in empirical models as a simple control variable. However, previous literature has already provided evidence towards the existence of a gender gap in a domain that is not so different from misperception, namely, political knowledge. This literature concluded that, ceteris paribus, women are less knowledgeable about political facts than men. Our work aims at shitfing the focus from political knowledge to the realm of erroneous perceptions: by employing cross-national survey data coming from the Eurobarometer and the European Social Survey, the contribution of our paper is twofold. First, we will test whether the gender gap also extends to misperceptions. Second, by interacting individual and contextual levels, we will show whether the gender gap in misperceptions is higher in countries with lower gender equality. To measure gender equality at the country level, we will employ the index recently developed by the European Institute for Gender Equality (EIGE).

Attitudes toward immigration and the effectiveness of state institutions
Federico Vegetti, Paolo Segatti
AbstractIn recent years, immigration has become one of the central issues of the political debate in many Western democracies. Given the importance attributed to immigration in shaping political preferences at several recent elections, it is crucial to understand the factors affecting citizens' opinions in this domain. However current individual explanations of immigration attitudes, largely focused on out-group attitudes or perceived sociotropic threat, tend to ignore people's beliefs about the state's ability to effectively manage migratory inflows, turning a potential threat into an opportunity. We theorize that an important factor affecting individuals' attitudes toward immigration is their trust in state institutions. We derive observable implications of our theory by looking at corruption, as a proxy for state functioning, and the overall effectiveness of state institutions. We test our expectations using comparative cross-sectional data from the Eurobarometer and the European Social Survey. Preliminary findings show that higher degrees of corruption in a country, as well as higher perceived corruption by individual respondents, are associated with their tendency to perceive immigration as a problem rather than an opportunity.

Trust in Political Institutions in Europe. A Cross-national Analysis using a Multilevel Approach
Isabella Mingo, Maria Paola Faggiano
AbstractTrust is an essential element for effective social cohesion and a correct expression of the principles of representative democracy. In many countries of the western world, we are witnessing the advance of anti-establishment (especially right-wing) political movements and a steady decline of trust in institutions. Crisis, perception of the crisis, climate of opinion around it have an impact on political attitudes and diversified feelings (interest, trust, resentment), on political participation and on voting (Morlino & Raniolo 2017; Diamanti & Lazar, 2018; Bordignon, Ceccarini & Diamanti, 2018). Therefore, a study on trust is relevant because investing or not on this resource has direct consequences on the structure and identity of a given community. Social analysts are called to deepen the reasons for the variability of these phenomena according to socio-cultural contexts and individual characteristics. The main aim of this paper is to analyse; 1) if and how much the spread of trust in institutions is different among the EU citizens and 2) what are the determinants of trust, deepening the intensity and the direction of the associations with subjective and objective indicators (Dogan, 2005; van der Meer, 2010). Our purpose is to outline the differences among the EU countries regarding trust in national and international institutions, considering at the same time macro (country) and micro (individual) level of analysis by using a multilevel approach. The empirical analysis is based on the latest data collected by European Social Survey (ESS-Round 8) in 2016 (microdata) and on some variables (macrodata) taken from other statistical sources (Eurostat, Transparency International). The sample of individuals counts about 40 thousand people interviewed face to face. According to the literature on this issue, the interpretative keys we chose are: socio-demographic and economic characteristics, political sophistication, cultural and political identity, satisfaction with own country, satisfaction with own life. At macro level we considered countries socio-economic features. Data analysis includes the following three steps: 1) the construction of synthetic indices; 2) the study of different levels of trust both at micro and macro level, applying descriptive techniques 3) identifying the determinants of trust in institutions, considering both micro and macro level explanatory variables, resorting to Multilevel Regression (Hox, 2010, Raudenbush & Bryk, 2002).

What Is In a Random Effect? The Role of Measurement Equivalence in Multilevel Analysis
Marco Steenbergen
AbstractWith the widespread availability of cross-national surveys and appropriate data analytic tools, in the form of multilevel analysis, comparative research of political behaviour has made great advances. One aspect that continues to receive scant attention, however, is the question of whether equivalent measures of constructs can be obtained in cross-national surveys. The answer to this question holds important implications for the data analytic tool of multilevel analysis. In this paper, I show that a lack of measurement equivalence results in biased estimates of variance components. Moreover, the interpretation of such components becomes much more difficult. Do they capture true structural differences in the effects of covariates across countries or do they merely pick up on differential item functioning? Having demonstrated the problem mathematically and through simulations, the paper then shows new types of multilevel analysis that can bypass the problems. Those methods are applied to data from the European Social Survey.