DEPARTMENT OF POLITICAL SCIENCE CENTRE FOR EUROPEAN STUDIES (CES) THE ROLE OF CULTURAL NARRATIVES AND SOCIAL IDENTITY IN PUBLIC SUPPORT FOR EUROPEAN ENVIRONMENTAL POLICIES. A survey experiment in the car-state Germany. Alicia Rathey Master’s thesis: 30 credits Programme: Master’s Programme in European Studies Level: Second Cycle Semester year: Spring 2025 Supervisor(s): Frederik Pfeiffer and Niklas Harring Abstract This thesis examines whether and how cultural narratives influence public support for environmental policies of the European Union (EU). It contributes to the field of European Studies by demonstrating how such narratives can shape the acceptance of EU policies within a culturally diverse union. Focusing on Germany’s car-state narrative, the study examines public support for the ban on internal combustion engine vehicles by 2035. Drawing on Social Identity Theory, the thesis explores how different framings of the policy activate identity- related mechanisms that shape individual support. To test these effects, a face-to-face survey experiment was conducted in Germany in April 2025 (N = 264). Participants were randomly assigned to two mini-experiments. The first group was exposed to an identity-affirming frame aligning with the cultural narrative of the car-state (TG1). The second group was exposed to an external threat frame presenting the policy as an EU mandate (TG2). The results show that aligning the policy with Germany’s car-state narrative increased support. A majority of participants in TG1 (93.9%) approved the ICEV ban when e-fuels were included as an option or held a neutral stance. This indicates that national cultural narratives can meaningfully shape public support for EU environmental measures. While identity-affirming frames clearly increase support, the EU frame, which was intended to trigger out-group threat, did not significantly reduce support as previously assumed. By demonstrating how national identity can be activated through targeted framing, the thesis contributes to a better understanding of the cultural conditions under which EU environmental policies gain public support. Master’s thesis: 30 credits Programme: Master’s Programme in European Studies Level: Second Cycle Semester year: Spring 2025 Supervisor(s): Frederik Pfeiffer and Niklas Harring Cultural narrative, public support, survey experiment, European Keyword: environmental policy, ICEV, Social Identity Theory, Germany Word count: 14 265 Contents List of Figures List of Tables List of Abbreviations 1. Introduction ........................................................................................................................ 1 1.1. Aim and research questions ................................................................................................... 3 1.2. Outline ................................................................................................................................... 4 2. Theory and previous research ............................................................................................ 4 2.1. Public support research .......................................................................................................... 4 2.2. Cultural narrative of a car-state .............................................................................................. 9 2.3. Research gap and contribution ............................................................................................. 11 3. Social Identity Theory and hypotheses ........................................................................... 11 4. Empirical framework and methodology ........................................................................... 15 4.1. Research design: survey experiment .................................................................................... 15 4.1.1. Operationalization of the concept ................................................................................................. 16 4.1.2. Justification and experiment standards .......................................................................................... 18 4.1.3. Structure of the survey experiment ............................................................................................... 20 4.1.4. Sample and data collection ............................................................................................................ 23 4.2. Limitations .......................................................................................................................... 26 5. Data analysis and results .................................................................................................. 27 5.1. Cultural narrative frame ....................................................................................................... 28 5.2. EU frame .............................................................................................................................. 30 5.3. Randomization ..................................................................................................................... 32 6. Discussion ....................................................................................................................... 33 7. Conclusion ....................................................................................................................... 37 References ................................................................................................................................ 38 Appendix .................................................................................................................................. 47 List of Figures Figure 1: Boxplot, TG1 vs CG1…………………………………………………….…………29 Figure 2: Boxplot, TG2 vs CG2…………………………………………………….…………31 List of Tables Table 1: Framing conditions and manipulation check applied to each group……………….…22 Table 2: Descriptive statistic, TG1 vs CG1……………………………………………………28 Table 3: Descriptive statistic, TG2 vs CG2……………………………………………………30 Table 4: Randomization check, Fischer’s exact test…………………………………………...32 Table 5: Linear regression analyses……………..………………………………………..…...33 List of Abbreviations BMW – Bayerische Motoren Werke CG – Control group EU – European Union EGD – European Green Deal EV – Electric vehicle ICEV – Internal combustion engine vehicle NOCS – number one car-state NOMS – number one mobility state SIT – Social Identity Theory TG – Treatment group VW – Volkswagen 1. Introduction Climate change is widely recognized as one of the most pressing challenges of the 21st century (Goerg et al., 2025, p. 1). However, it is not only an environmental issue but also a deeply politicized one, with societal divisions emerging over causes, responsibilities and appropriate policy responses (Klas et al., 2022, p. 2). To better understand these societal divisions over environmental policies, it is crucial to examine how cultural narratives shape public support for environmental measures. This is particularly relevant in the context of the European Union’s (EU) ambitious climate agenda aimed at achieving climate neutrality. The European Council emphasizes that „EU countries are legally committed to fighting climate change by shifting to a climate-neutral economy with net-zero greenhouse gas emissions by 2050” (European Council, 2025). The European Green Deal (EGD) serves as the strategic framework for this transformation, introducing comprehensive measures across multiple sectors, including transport (European Commission, n.d.). Against this background, this thesis investigates how Germany’s cultural narrative as a car-state influences public support toward EU environmental policies1, specifically focusing on the 2035 ban on internal combustion engine vehicles (ICEV) as a climate policy example. The transport sector is central in this context because it accounts for 25% of the EU’s greenhouse gas emissions, with road traffic responsible for 71.7% of these emissions. Breaking it down further, 60.6% of these emissions come from car use, making passenger vehicles a primary contributor. This sector is also the only one where emissions have increased by 33.5%, despite overall climate efforts. Addressing transport emissions is therefore critical for achieving the EU’s climate goals (European Parliament, 2024a). Recognizing this challenge, the European Commission, European Parliament and European Council agreed in 2022 to phase out the sale of new ICEVs by 2035 (European Commission, 2022). This landmark decision demands profound structural changes from Europe’s automotive sector, an industry of significant economic and symbolic importance. However, early indicators suggest that this transition faces challenges. Electric vehicle (EV) sales declined by 10% across Europe and by a striking 37% in Germany (European Parliament, 2024b). At the same time, global competition, particularly from China and the United States, has intensified, further pressuring European manufacturers (van Wieringen, 2024, pp. 8, 16-17). 1 In this thesis, the terms climate policies and environmental policies are used synonymously. Strictly speaking, climate policies focus on mitigating climate change and reducing greenhouse gas emissions, while environmental policies encompass a broader range of environmental issues. However, as climate policies are considered a subset of environmental policies, no further differentiation is made in the following (Dhage, 2024, pp. 129, 131; European Parliament, 2024d). 1 The economic relevance of the automotive industry varies across EU member states and regions. In countries such as Slovakia, Romania, Sweden, Czech Republic, Hungary and Germany, over 10% of all manufacturing jobs are directly linked to the industry (European Parliament, 2024c). Beyond its economic importance, the automotive industry carries significant symbolic value. Examples like Made in Germany or Italian car design stand for European innovation and craftsmanship (Cornet et al, 2023, p. 2). Germany holds a particularly prominent position. As the EU’s largest economy with a historically export-driven industrial model, its automotive sector is not only economically important but also deeply embedded in the national identity (Haas, 2020, p. 4). Automobility is symbolized by brands like Volkswagen (VW), Mercedes-Benz and Bayerische Motoren Werke (BMW) as well as internationally well-known infrastructures like the Autobahn. Consequently, the car-state narrative has become a collective self-image as a leading engineering nation with a strong automotive industry. This cultural narrative has historical roots reaching back to the postwar economic miracle and remains powerful today, shaping resistance to change, particularly when proposed policies threaten this established self-image (Kunze, 2022, pp. 13, 25). Germany’s political influence further amplifies the significance of this narrative. Although Germany was initially supportive of the EGD regulations, it unexpectedly withheld approval during the final formal vote in the Council of Transport Ministers. The country insisted on an exception for vehicles running on synthetic e-fuels (Birel et al., 2024, p. 1). E-fuels are synthetic fuels produced using renewable energy, water and carbon dioxide. They are considered a potential carbon-neutral alternative to conventional fossil fuels, enabling the continued use of existing ICEVs2 (ADAC, 2024). This suggests that, for the time being, there is no need for Germany to make any changes. However, Germany’s policy reversal caused more opponents of the law to reverse their stance (Birel et al., 2024, pp. 1, 6). This political maneuver activated a broader debate and led to a compromise. While the ICEV ban remains in place, it now includes an exception for ICEVs powered by e-fuels (Bundesregierung, n.d.). The German insistence on protecting traditional ICEV technology can be interpreted as more than an economic strategy, it reflects a broader cultural attachment to automobility. As Social Identity Theory (SIT) (Tajfel & Turner, 2004) posits, individuals derive a sense of self-esteem and belonging from their group memberships. When policies are perceived as threatening core 2 Hydrogen cars are another alternative technology to support climate-neutral mobility. They use a fuel cell that converts hydrogen into electricity through a chemical reaction with oxygen. Therefore, the car is powered by an electric motor. In contrast, e-fuels are burned in a conventional ICEV like gasoline or diesel (Carbo Europe, 2025). 2 elements of a group’s identity, resistance can be emotional and symbolic, not merely rational or interest-based. Therefore, the resistance to environmental policies can be identity-driven. 1.1. Aim and research questions The aim of this thesis is to explore how national cultural narratives influence public support for European environmental policies. Using Germany as a case study, it investigates how the cultural narrative of Germany as a car-state shapes public support toward the EU’s climate initiatives, with a particular focus on the 2035 ICEV ban. While cultural narratives are often latent within collective identity, they may become salient when strategically activated through framing. Therefore, this thesis further examines how different framings that reference the car-state narrative can either reinforce or mitigate public resistance to the policy. Beyond addressing the link between culture and support, the thesis seeks to provide experimental evidence on how identity-based factors can act both as barriers to and as opportunities for promoting sustainable transitions. In doing so, it contributes to existing research by explicitly considering cultural narratives as a determinant of public support for climate policies. Therefore, the thesis asks the following research questions: (1) Does the cultural narrative of Germany as a car-state influence public support for European environmental policies and (2) how does the framing of the narrative influence the support? The choice of Germany as a case study is particularly compelling for three key reasons. First, Germany’s insistence on including e-fuels in the EU’s climate strategy can be seen as a political maneuver to preserve the ICEV, reflecting the country’s deep-rooted automotive identity. Second, it also underscores Germany’s influential position within the EU, where it can shape climate policies according to its national interests. Third, Germany simultaneously faces increasing pressure to decarbonize its transport sector and uphold its climate commitments. This tension, between a culturally embedded car-state identity and ambitious environmental goals, makes Germany an ideal case for exploring how national cultural narratives influence public support for climate policies, particularly those initiated by supranational institutions like the EU. 3 1.2. Outline The thesis is structured as follows. Chapter 2 provides an overview of relevant theory and previous research, focusing on public support for climate policies and the role of cultural narratives. It concludes by outlining the research gap and the specific contribution this thesis aims to make. Chapter 3 introduces the theoretical framework of SIT, explains its relevance to the research context and presents the derived hypotheses. The following chapter (chapter 4) describes the empirical framework and methodology in detail, including a comprehensive explanation of the survey experiment and the key concepts necessary for its design. It also reflects on the limitations of the chosen approach. Chapter 5 presents the data analysis and results, followed by a discussion (chapter 6) that interprets the findings in relation to existing literature, provides broader reflections and suggests directions for future research. Finally, chapter 7 concludes by summarizing the key insights and answering the research questions. 2. Theory and previous research This chapter begins with a review of the existing literature on public support for environmental policies, outlining key findings and identifying the research gap addressed by this thesis. It then introduces the concept of cultural narratives, with a focus on their role in shaping collective identity. Particular attention is given to the narrative of Germany as a car-state. 2.1. Public support research Public support can be understood as the general public’s approval of government measures, whether proposed or already implemented, that are intended to mitigate (environmental) issues, including climate change (Heyen & Wicki, 2024, p. 786). Closely related concepts include public opinion, which is a significant determinant of policy change, particularly in democratic countries (Drews & van den Bergh, 2015, p. 855). Another important concept is acceptability, which is defined as an affirmative but passive evaluative response toward a policy proposal that may or may not lead to supportive or opposing behavior (Heyen & Wicki, 2024, p. 786). Public support can be expressed in various forms, such as agreement in surveys, a willingness to incur personal costs or electoral behavior, whether in favor of or against specific measures (Fairbrother et al., 2021, Bergquist et al., 2022). Public support research refers to the study of how and why individuals or groups express approval. It aims to understand the factors that shape support or opposition among the public, such as personal values, political ideology, trust in institutions, perceived fairness and effectiveness of policies or socio-demographic 4 characteristics (e.g. Drews & van den Bergh, 2015; Bumann, 2021; Bergquist et al., 2022, Fairbrother, 2022). Understanding public support is vital, as the level of public support is a critical factor in the successful implementation of climate policies (Drews & van den Bergh, 2015, p. 855; Baute, 2024, p. 3). Without a clear understanding of what people support and why, policies risk political failure (Fairbrother, 2022, pp. 1, 6). Although public support is not the only factor influencing political decisions (e.g. interests, resources and strategic behavior of political actors), it remains a key element in the political decision-making process (Heyen & Wicki, 2024, p. 790). Policymakers may hesitate to implement climate policies if they anticipate public opposition (Drews & van den Bergh, 2015, p. 855). Conversely, public pressure can also accelerate sustainability transitions (Hoppe et al., 2023, p. 821). Hence, Nilsson and Weitz (2019, pp. 255, 259) emphasize that managing trade-offs and conflicts constitutes the core of political decision-making and necessitates negotiation, a reality that leads environmental policy analysts to anticipate trade-offs between problem-solving effectiveness and political feasibility. Therefore, understanding the determinants of public support can help guide policymakers in their policy choices to ensure successful implementation (Bergquist et al., 2022, p. 235). Public support for environmental policies has been extensively studied and is widely recognized as being influenced by a variety of factors (e.g. Anderson et al., 2017; Bergquist et al., 2022; Nielsen et al., 2024). Consequently, a substantial body of literature reviews already exists, synthesizing empirical findings and offering theoretical frameworks for understanding the determinants of support. A notable contribution is the cross-disciplinary review by Drews and van den Bergh (2015, p. 855). The authors state that, to their knowledge, it is the first review on the topic making it a conceptual groundwork (ibid, p. 856). Therefore, it is not surprising that the review has been cited multiple times, including by Bumann (2021), Fairbrother (2022) and Heyen and Wicki (2024). The authors note that a substantial body of empirical literature had emerged in the years prior to their publication, approaching the subject from different angles (Drews & van den Bergh, 2015, p. 856). They categorized factors influencing public support into social- psychological factors, perception of policy design and contextual factors. This categorization provides a useful framework for understanding the determinants of policy support (ibid, p. 855). Their review highlights that left-wing political orientation, environmental values, perceived policy effectiveness, fairness and social trust positively influence support (ibid, p. 867). Moreover, they noted the preference for pull over push measures (ibid, pp. 859). Importantly, 5 Drews and van den Bergh (2015, p. 869) emphasize the need for more experimental research to establish causal relationships and call for broader investigations across diverse geographical and cultural contexts. Similarly, Bumann (2021, pp. 214-215) conducted a comprehensive review focused on empirical studies using polling and survey data. Similar to Drews and van den Bergh (2015), he notes the field’s diversity, particularly in terms of concept, policy support measures and empirical approaches (Bumann, 2021, p. 222). Bumann’s review covers public support for broad climate actions, specific policy instruments (economic and regulatory), renewable technologies and index measures of policy support. The review mainly examined the extent and the determinants of individuals support for different climate policies from these various perspectives. His main argument is that public support is primarily driven by climate change beliefs and party identification rather than socio-demographic factors (ibid, p. 213). Bumann further underscores the importance of accounting for regional differences in public opinion when analyzing or designing policies (ibid, p. 224). Building on these insights, Fairbrother (2022, pp. 1-3) offers a more detailed discussion of public support for environmental policies, with a particular focus on climate policies such as carbon taxes. His review emphasizes that belief in climate change does not automatically translate into support for climate policies, pointing to a crucial gap between environmental concern and policy acceptance (ibid, p. 4). Fairbrother further identifies the design of the policy itself and the strategies used for communication as critical factors for increasing public acceptability. In particular, he highlights the need for further research into how to appeal to individuals who are otherwise skeptical of climate policies, how variations in policy features influence popularity and how different types of framing might enhance support (ibid, p. 9). In contrast, Heyen and Wicki (2024, pp. 785-787) offer a distinct, critical perspective by challenging the dominant focus on individual-level and attitudinal factors. Their main contribution is the concept of governable acceptability factors, aspects of policy design that lie within the control of policymakers, such as the timing of implementation, inclusion of stakeholders and communication style. They argue that previous research has focused too heavily on variables that are difficult or impossible to influence through policy, such as ideology or deeply held values (ibid, pp. 787-788). They state that more research is needed on these factors to help design and enforce more acceptable stringent climate policies (ibid, p. 790). Together, these four reviews were selected not only because they represent different methodological and disciplinary perspectives, but also because they collectively reflect the 6 development and current state of the literature on public support for environmental policies. Their combination enables a multi-layered understanding of what drives public support and where the field still falls short. Despite their breadth, all four reviews converge on several limitations that are directly relevant to this thesis. First, there is widespread agreement on the lack of experimental research capable of establishing causal effects (Drews & van den Bergh, 2015; Fairbrother, 2022). Second, the unexplored impact of framing and policy design that can be actively shaped by policymakers to improve public support (Fairbrother, 2022; Heyen & Wicki, 2024). Third and lastly, the need for regional differences in public opinion when analyzing or designing policies (Bumann, 2021). These gaps underscore the need for experimental designs that test how concrete elements of framing influence public support, particularly in different cultural contexts. Therefore, Germany presents a compelling case for further examination. As a country with a strong historically grown automotive identity, Germany offers a unique context in which cultural narratives interact in distinct ways (see chapter 2.2.). Recent research focusing specifically on Germany offers valuable insights into how public support for European environmental policies manifests nationally. Baute (2024, pp. 1-2) investigated public support for European climate policies among the German population. He found that, while the majority acknowledge anthropogenic climate change and anticipate negative consequences, the design of specific policies remains crucial for public acceptance. Only a small minority (less than 6%) either reject climate change or attribute it solely to natural causes (ibid, pp. 3, 6). Building on this, Georg et al. (2025, p. 2) demonstrated that public support for specific, ambitious policies can be fragile. Consequently, ambitious climate policies reduce public support in Germany. Particularly, carbon prices reduce public support more significantly than policy mixes focusing on greenhouse gas reductions. Their regression analysis reveals that policy support is also significantly associated with economic preferences and individual characteristics, as well as regional characteristics. Notably showing stronger declines in support in East Germany (ibid, pp. 6-8) and highlighting the complexity of policy support measures mentioned before. Similarly, Mögele and Rau (2020, p. 17) highlight regional differences, noting lower support in South Germany, where the automotive industry’s historical presence reinforces attachment to car-centric identities and practices. Turning to the transport sector, specifically, achieving decarbonization goals requires not only adopting net-zero technologies but also phasing out greenhouse gas-emitting technologies. However, implementing policies in this sector can have a direct and substantial impact on 7 people’s daily lives, potentially leading to public opposition. This argument has been put forward by Tröndle et al. (2023, p. 6). The authors further explain bans are preferred for heating systems in the building sector, while economic instruments like taxes are preferred for ICEVs in the transport sector. Showing that policy support differs by sector (ibid). Hoppe et al. (2023, p. 824) confirm that in the case of an ICEV ban pull measures such as public transport or EV purchase subsidies receive higher support than bans. Nevertheless, although 45% of the German public supports the general idea of an ICEV phase-out, concrete policy proposals only achieve 29% approval (ibid, p. 819). Hoppe et al. further reveal that emotional attachment to automobility remains strong. Meaning, that fears about negative impacts on national car manufacturers, personal mobility and the economy are prevalent. Additionally, the actual EU’s proposal to ban had little effect on the answers (ibid, p. 823). Tröndle (2023) links this (sectoral) resistance to Germany’s automotive cultural narrative, suggesting that attachment to automobility influences public responses to transport policy measures. Complementing these findings, Bergfeld et al. (2023) explored user acceptance of e-fuels for passenger vehicles. The survey found that usage behavior, gender and environmental behavior have an influence on fuel choice and acceptance of e-fuels (ibid, pp. 1308-1309). The authors state that the results should be understood as a directional guide due to limitations like low current availability of e-fuels (ibid, p. 1309). All three transport-sector-focused studies emphasize Germany’s relevance as a case study, particularly given its powerful car industry and strong automotive identity. While they recognize the importance of cultural narratives as background explanations for public support, none of these studies directly operationalizes cultural narratives as influencing factors. This gap further motivates this thesis, which seeks to integrate cultural narratives explicitly into the experimental design. In addition to policy support, research also points to the crucial role of communication and framing. Drexler et al. (2022) and Mögele & Rau (2020) delve into the complexities of Germany’s transport sector transformation, particularly challenging the prevailing dominance of automobility. While both studies acknowledge the identity of a car-state (Mögele and Rau, 2020) and the complexity of the Verkehrswende (mobility transition) (Drexler et al., 2022), they approach the topic with different theoretical frameworks. Mögele and Rau (2020, pp. 18-19) examined two different concepts as prominent storylines used in public communication surrounding the future of the automobile industry and mobility in Southern Germany, particularly in Bavaria and Baden-Württemberg. Using an argumentative 8 discourse analysis they analyzed how actors (re-)construct mobility-cultural meanings and practices through their communication and action (ibid, p. 16). One storyline is the number one car-state (NOCS) and the other is the number one mobility state (NOMS) (ibid, p. 22). The NOCS narrative reinforces the historical legacy of automotive dominance. Conversely, the NOMS narrative aims to broaden the regional identity to encompass the entire mobility system. However, political negotiations often reframe NOMS to still include car manufacturing, leading to a cars and mobility rather than cars versus mobility interpretation. This dynamic results in a paradoxical reinforcement: Efforts to promote transformation end up stabilizing the car- centered identity, driven by fears of economic disruption, social upheaval and restrictions on individual freedoms (ibid, pp. 23-25). Drexler et al. (2022, pp. 1-2) analyzed the framing of the Verkehrswende in Germany by various actors (industry, science, politics, media), integrating the Multi-Level Perspective on socio- technical transitions with an interdisciplinary framing approach. Their framing analysis focuses on problem descriptions, causal attributions, moral evaluations and proposed interventions (ibid., p. 3). They find that, while most actors agree that a transition is necessary, significant diversity exists regarding how problems are defined and what solutions are proposed. Industrial actors and governing parties often favor improving propulsion technologies, whereas left-wing parties, scientists and media actors advocate for fundamental behavioral and structural shifts away from private car dependence (ibid, p. 12). Both studies demonstrate that there is a shared understanding of the necessity of a transformation. However, Drexler et al. (2022) show that there remains no common understanding of the underlying problems or appropriate solutions, while Mögele and Rau (2020) argue that current efforts paradoxically reinforce the dominance of the existing automobility culture. Their work highlights the critical importance of recognizing and addressing cultural identity formation processes within transition governance if genuine systemic change is to be achieved. 2.2. Cultural narrative of a car-state To understand how cultural narratives shape public support for climate policies in Germany, it is essential to clarify the specific form such narratives take. While Mögele and Rau (2020) highlight the relevance of automobility-related storylines, they do not explicitly define what constitutes a cultural narrative or what they precisely mean by the notion of a car-state. This chapter therefore aims to conceptually specify the car-state as a cultural narrative to make it analytically usable. 9 Cultural narratives refer to the stories that societies construct and pass down to maintain a sense of shared history, identity and collective memory. These narratives shape how historical events are remembered, interpreted and framed (Müller-Funk, 2007; Manderscheid, 2024). Müller- Funk (2007, p. 251) emphasizes that cultural narratives serve as identity-forming structures that help societies make sense of their past and present. This understanding of cultural narratives is supported by Historian Conrad Kunze (2022), who argues that enthusiasm for the automobile is deeply rooted in German history, extending far beyond its economic relevance. He describes Germany’s relationship with automobility as a coexistence with both the Auto and the Autobahn (car and the German highway). For instance, the Autobahn is not merely an infrastructure project, it has been mythologized as a symbol of modernity, freedom and national identity (ibid, p. 18). Kunze (2022, pp. 303-312) further illustrates how the postwar economic miracle solidified the car as a central pillar of German industrial success. Brands such as VW, Mercedes-Benz and BMW became global symbols of German engineering excellence, reinforcing the automobile’s place in the national consciousness. This aligns with Müller-Funk’s (2007, p. 252) concept of cultural narratives as bridges between memory and identity, helping societies reframe historical events in ways that support national cohesion. However, while postwar Germany distanced itself from many aspects of its fascist past, the ideological significance of the automobile remained largely unquestioned. Kunze (2022, p. 40) highlights how debates about the speed limit have been politically instrumentalized since the Nazi era, with slogans such as free travel without a speed limit being used for electoral campaigns. This underscores how transport policy is deeply entangled with identity politics, making car-related regulations highly contentious. Additionally, the Autobahn captures the industrial success as a symbol (e.g., an important symbol for jobs since National Socialism). While the thesis focuses on the ICEV ban, the Autobahn remains relevant because it embodies the cultural attachment to driving without a legally enforced speed limit and high-performance vehicles, a key aspect of Germany’s car culture (ibid, p. 25). According to Müller-Funk (2007, pp. 251-252), cultural memory plays a crucial role in the formation of individual and collective identities. This highlights the paradox of cultural narratives. They claim to represent the past but are always shaped by present concerns. In the case of Germany’s car-state identity, this paradox becomes evident. The narrative of automobility continues to influence national debates on mobility, environmental policies and identity, revealing deeply rooted tensions between tradition and transformation. 10 For the purposes of this thesis, Kunze’s (2022) conceptualization of the Auto is used to summarize the car-state narrative. Auto represents economic strength and engineering excellence as part of Germany’s identity and global reputation. 2.3. Research gap and contribution The impact of cultural factors on public support, especially in the context of Germany’s automotive identity, remains insufficiently explored. Studies that do touch on cultural factors, mostly in combination with stakeholders and political actors, such as Mögele and Rau (2020) and Drexler et al. (2022), focus on qualitative analysis, revealing a need for quantitative research that systematically measures public opinion. Moreover, these studies often lack a precise conceptualization of what cultural narratives entail and how they influence support toward environmental policies. This thesis seeks to close this gap by applying the concept of cultural narratives. Using Germany’s car-state narrative, it investigates whether and how this influences public support for the 2035 ICEV ban policy. To empirically capture this effect, the study uses different framings of the policy in an experimental design to activate identity-relevant aspects of the cultural narrative and measure their influence on policy support (see chapter 4). In doing so, this research provides empirical insights into the cultural foundations of policy support and responds to calls for more context-sensitive and experimental approaches in the field. 3. Social Identity Theory and hypotheses This chapter presents the theoretical framework guiding this thesis, as well as the developed hypotheses based on Social Identity Theory (SIT). SIT provides an analytical framework through which to understand how collective identities, such as religions, communities or nations, influence public support (Dubois, 2001, p. 2214; Hornsey, 2008, p. 204). In the context of this thesis, it is used to make the cultural narrative salient. SIT posits that individuals develop a significant part of their identity from the social groups to which they belong and that the desire to maintain a positive social identity influences intergroup attitudes and behavior (Tajfel & Turner, 2004, p. 284). SIT emerged in 1978/9, primarily through the work of Henri Tajfel and John Turner (Trepte & Loy, 2017, p. 1). SIT was developed as a response to what Tajfel and Turner perceived as overly individualistic and reductionist tendencies in existing theories of intergroup relations (Tajfel & Turner, 2004, p. 276; Hornsey, 2008, p. 205). Their initial work involved the famous minimal group paradigm. In these experiments, participants were arbitrarily categorized into groups 11 based on trivial and meaningless criteria, such as estimations of dots on a page or the flip of a coin (Tajfel & Turner, 2004, pp. 281-282; Hornsey, 2008, p. 205). Despite the minimal basis for group affiliation, participants consistently showed in-group favoritism, allocating more points or resources to members of their own group compared to the out-group (ibid; Mangum & Block, 2018, p. 3). These findings suggested that intergroup bias could arise simply from the act of social categorization itself, even in the absence of realistic conflict or prior animosity. Tajfel and Turner (2004, p. 284) argued that this demonstrated a fundamental human motivation to positively evaluate one’s own group. The SIT approach gained broader international attention in the 1980s and 1990s and has since become a cornerstone in the study of group dynamics (Hornsey, 2008, p. 205). SIT is built upon several key interrelated concepts. Social identity refers to the part of an individual’s self-concept that derives from their membership in a social group together with the value and emotional significance attached to that membership. Individuals strive to maintain a positive social identity because it contributes to their overall self-esteem (Trepte & Loy, 2017, p. 5). Social categorization means the cognitive process of classifying individuals (including oneself) into social groups based on shared characteristics such as nationality, profession or beliefs. Social categorization simplifies the social world, allowing for efficient information processing. It is an automatic process where people naturally think in group terms (ibid, p. 3). With that, the concept of social identification describes that once individuals categorize themselves as belonging to a particular social group, they begin to identify with that group. This involves adopting the characteristics, norms and values of the in-group. Social identification is a psychological attachment to the group. The more strongly an individual identifies with a group, the more their thoughts and behaviors will be influenced by their group membership (ibid, pp. 5-6; Taifel & Turner, 2004, p. 283). These concepts are demonstrated in in-group vs. out-group. The in-group is the social group with which an individual identifies as a member (“us”), while the out-group is a social group with which the individual does not identify (“them”) (Mangum & Block, 2018, pp. 3,5). The minimal group paradigm demonstrated that even arbitrary distinctions can lead to the perception of an in-group and an out-group (Horsey, 2008, p. 205). This occurs particularly when the group identity is perceived to be under threat (Mangum & Block, 2018, p. 10). Following categorization and identification, individuals engage in social comparison between their in-group and relevant out-groups. The primary goal of this comparison is to establish the in-group as positively distinct from the out-group (Trepte & Loy, 207, pp. 3-4). Collectively, the concepts contribute to positive distinctiveness, which describes the motivation to view one’s own group as superior or more legitimate than others. 12 Meaning, group affiliation becomes more than social belonging, it becomes a critical component of self-concept and a driver of behavior in identity-relevant contexts (ibid, p. 4). SIT is particularly well-suited for this thesis because it captures the affective dimensions of group identity. In Germany, automobility is not merely a key economic sector but also a central element of national identity. The mythology of the Autobahn, the global reputation of German car brands and the car’s role as a symbol of success and pride, all contribute to a shared narrative that reinforces collective belonging (see chapter 2.2.). As Mangum and Block (2018, p. 6) argue, individuals tend to affiliate with groups that hold high status and the German automotive identity has long been one of prestige. When this identity is perceived to be under threat, for example from EU policies, such interventions may be interpreted not as neutral policy shifts but as symbolic attacks on core values. Barnett et al. (2021, p. 1) emphasize that social identity is maintained through the perception of continuity, distinctiveness, self-esteem and self- efficacy. Environmental regulations that challenge the dominant car-state narrative, such as the ban on ICEVs, may disrupt all four of these dimensions. For example, perceived loss of technological distinctiveness or national autonomy in transportation policy can provoke resistance not because of the policies themselves, but because they are experienced as identity threats. The car-state narrative therefore functions as a psychological anchor, reinforcing group boundaries and contributing to a sense of in-group cohesion. In contrast, groups and individuals that promote environmental reforms are often perceived as outsiders or out-groups. Furthermore, SIT is particularly relevant when examining public responses to issues that are perceived to have implications for the standing and distinctiveness of important social group. When policies or proposals are framed in ways that suggest a threat to the in-group’s values, norms, resources or overall positive distinctiveness, individuals may react defensively to protect their social identity (Mangum & Block, 2018, pp. 3, 13-14). Therefore, SIT offers a useful framework for interpreting how Germans may respond to European environmental policies and especially the ICEV ban. From this perspective, when such policies are framed in ways that activate the in-group identity or are perceived as threatening its distinctiveness, several intergroup dynamics are likely to be triggered. For instance, individuals who strongly identify with the cultural narrative are likely to display in- group favoritism and support policies that appear to protect the interests of the in-group (Hornsey, 2008, p. 207). Meaning, that individuals with a strong group identity are less likely to support the ICEV ban. However, introducing e-fuels as an alternative can mitigate this resistance by preserving key symbolic elements of the traditional ICEV and the cultural 13 narrative. It is not merely the physical car that holds cultural significance, but the entire experience associated with it, from owning an ICEV, day to day activities like refueling at gas stations, to enjoying high-speed driving on the unrestricted Autobahn (Mögele and Rau, 2020, pp. 15, 24; Manderscheid, 2020, p. 45). Conversely, when EU environmental policies are perceived as a threat to this identity, out-group derogation may occur (Hornsey, 2008, p. 207). This could manifest as negative support toward actors associated with the policy threat, not only directed at the EU in general but also green parties or environmental activists (Tajfel & Turner, 2004, p. 283). Additionally, individuals may respond to identity threats by engaging in efforts to restore positive distinctiveness (Tajfel & Turner, 2004, p. 283). This could include emphasizing Germany’s potential as a leader in green automotive innovation or reaffirming the central economic and cultural role of the automotive sector. These psychological mechanisms illustrate how individuals may interpret policy proposals through the lens of group identity, seeking to defend a positive in-group image in the face of perceived external pressure (Trepte & Loy, 2017, p. 4). In this context, it is important to define the relevant in-group and out-group dynamics. The primary in-group consists of individuals who strongly identify with Germany’s cultural narrative. For these individuals, the cultural narrative is a key element of their social identity, contributing to their sense of belonging and national pride (→ H1a). In contrast, the out-group includes individuals who support European environmental policies that are perceived as threatening this established identity. While the EU itself is not inherently viewed as an out- group, policies framed as EU mandates may trigger defensive reactions when perceived as external impositions challenging national distinctiveness (→ H1b). Based on these considerations, this thesis hypothesizes that public support for EU environmental policies will vary depending on how such policies are framed. Specifically, policies framed to align with Germany’s cultural narrative are expected to increase public support by resonating with in-group identity and values. Conversely, policies framed as top- down EU mandates are expected to decrease public support, as they may be perceived as external impositions (out-group) challenging national distinctiveness. 14 The following hypotheses are derived from SIT: H1a: EU environmental policies framed to align with Germany’s cultural narrative will increase public support. H1b: EU environmental policies framed as an EU mandate will decrease public support. This thesis therefore not only applies SIT to better understand public support but also explores how framing interacts with identity defense mechanisms. As Barnett et al. (2021) demonstrate, identity salience can be triggered through strategic communication, suggesting that framing is a crucial mechanism for shaping perception and response. This supports the use of a survey experiment with different policy frames designed to activate distinct identity responses, an approach that will be outlined in detail in the following chapter. 4. Empirical framework and methodology This chapter presents the empirical framework and methodological approach used to test the thesis’s hypotheses. It begins by outlining the research design, which is based on a survey experiment and explains how key theoretical concepts are operationalized into measurable variables. The structure of the experiment, the treatment conditions and the rationale behind the policy framings are described in detail. The chapter also discusses the sampling strategy and data collection, followed by a reflection on methodological limitations. To ensure the validity and transparency of this thesis, particular attention is paid to the clarity, replicability and coherence of the chosen approach (Weimann & Brosik-Koch, 2019, p. 172). 4.1. Research design: survey experiment The aim of this thesis is to examine how cultural narratives influence public support for European environmental policies. Drawing on the SIT, this thesis assumes that public support for EU environmental policies is significantly influenced by the way in which these policies are framed in relation to Germany’s cultural narrative surrounding the automotive industry. Specifically, it is expected that framing an EU policy in a way that resonates with Germany’s national identity will increase public support. In contrast, policies framed primarily as EU mandates are hypothesized to elicit greater resistance, as they may be perceived as external interference, triggering in-group defensiveness and reducing acceptance. 15 In this thesis, cultural narratives function as the independent variable, while public support for EU environmental policies represents the dependent variable. However, because cultural narratives are deeply embedded in collective memory and cannot be directly manipulated in a short-term survey setting, the thesis uses framing as an experimental tool to make this narrative salient to the participants. The frame of an experiment is defined as the manner in which a specific decision-making problem is presented to participants (Weimann & Brosig-Koch, 2019, pp. 121-123). This approach follows Robert M. Entman’s (1993, p. 52) definition, that framing is “to select some aspects of a perceived reality and make them more salient in a communicating context, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation for the item described”. This definition guides the experimental manipulation, which involves presenting the same EU climate policy, the ban on ICEVs, in distinct frames, either aligned with the cultural narrative or emphasizing the EU mandate (see chapters 4.1.1. & 4.1.4.). To empirically test the hypothesized effects, the thesis uses a survey experiment, a method particularly suited to identifying causal effects in a controlled yet realistic setting (Rack & Christophersen, 2009, pp. 18, 31). The survey format allows participants to engage with short policy texts that vary only in framing, enabling a clean comparison. At the same time, the realism of the policy scenario ensures external validity, while random assignment to treatment conditions supports internal validity (Himme, 2009, p. 491). In the following subsections, the research design including a detailed description of the experiment is presented. 4.1.1. Operationalization of the concept To empirically test the proposed relationship between cultural narrative and public support, the key theoretical concepts must first be translated into measurable variables3. The dependent variable is public support for EU environmental policies, specifically the ban on ICEVs by 2035. One survey question was used to capture public support: What is your opinion on this policy?4 The responses were measured using a five-point Likert scale ranging from do not agree at all, rather disagree, neutral, agree somewhat and strongly agree. The Likert scale enables a 3 See chapter 4.1.3. (face-to-face survey experiment) for the final order and structure of the questionnaires. 4 All questions, scales and formulations are direct translations from German, as the questionnaire was conducted in German. As a result, some translations may sound unusual, but they are necessary to preserve the original meaning and content. 16 nuanced measurement of support and allows for an analysis of how the experimental conditions influence respondents’ support (Greving, 2009, pp. 73-74). The independent variable in this thesis is the cultural narrative employed as a framing condition, which constitutes the different treatments under which the dependent variable is measured (Weimann & Brosig-Koch, 2019, p. 226). The independent variable is manipulated experimentally through slight variations in the wording of the policy description (see chapter 4.1.3.). These variations present the same EU policy from different perspectives, either affirming or challenging elements of the cultural narrative. This approach enables investigations of how identity-relevant cues influence public support. From an experimental perspective, this distinction is crucial, because a variable can only be considered causal within an experimental design if it is actively manipulated by the researcher (Weimann & Brosig- Koch, 2019, pp. 226-227). Since a cultural narrative cannot be easily manipulated in a short survey setting, it is more appropriate to manipulate the frame in which a given policy is embedded. To measure a degree of randomization in the experiment, the survey included three socio- demographic variables: age, gender and education. These standard control variables help assess how policy support may vary across population segments (e.g. Bergfeld et al., 2023, p. 1305; Georg et al., 2025, p. 5). However, the main reason for including these control variables is to address potential omitted variable bias while avoiding overcontrol. This aligns with the best practices of experimental design, which emphasize simplicity and clarity (Mutz & Pemantle, 2015, p. 210). It ensures that potential confounding variables, whether observable and unobservable, are disrupted roughly equally across the different treatment groups (Weimann & Brosig-Koch, 2019, p. 228). Education is widely recognized as a significant factor in shaping environmental attitudes, with higher levels of education often linked to greater awareness of environmental issues and stronger support for policy measures (Harring & Jagers, 2017). Similarly, age plays a role in shaping environmental attitudes, as younger individuals tend to be more environmentally conscious due to generational shifts in values and greater exposure to environmental education (Drew & van den Bergh, 2016). Gender differences have also been observed, with research suggesting that women are generally more supportive of environmental policies than men. However, the extent of these differences varies across national contexts, highlighting the influence of broader social norms (Arora-Jonsson, 2014). 17 The variable education is measured using the following survey question: What is your highest level of education? The responses are then coded into one of the four categories: no school leaving certificate, general school leaving certificate, professional qualification and academic degree. The identification of the gender is measured with female, male, diverse or no answer. Age is categorized into four groups: 18–29 years, 30–49 years, 50–66 years and 67 years or older. Both the age and education variables were grouped this way to enable clear and easy statistical comparison and to ensure sufficient variation across treatment groups, contributing to the validity of the random assignment. To capture the behavioral and attitudinal components of automotive identity, two additional questions were included: Do you own a vehicle? (yes/no) and If yes, which type? were asked. The second question will be coded in three categories: internal combustion engine vehicles (petrol, diesel, gas, etc.), electric vehicles and hybrid vehicles. This follows the approach of Bergfeld et al. (2023, p. 1305), who used vehicle ownership and usage patterns to link real- world behavior with environmental attitudes. 4.1.2. Justification and experiment standards A survey experiment was chosen because it allows researchers to manipulate independent variables and measure their effects on dependent variables in a controlled setting (Mutz & Pemantle, 2015, p. 195). Unlike laboratory experiments that often focus on tangible, observable behaviors, survey experiments are well-suited for manipulating latent constructs such as attitudes, perceptions and social identities (ibid, p. 196). They have become an established method in research, particularly for studying framing effects, attitude change and cognitive processing among respondents (Weimann & Brosig-Koch, 2019, p. 55). Through the deliberate variation of frames, this method enables a more precise assessment of causal relationships than traditional observational designs. A face-to-face survey format was chosen to maximize response rates and secure the minimum number of participants necessary for representativeness. This format was particularly well-suited to the logistical and methodological constraints of a master’s thesis, allowing for flexibility in administration and direct engagement with respondents. In-person interaction also made it possible to ensure consistent comprehension of survey items, which is especially important in experimental conditions involving subtle language variations (Kaya, 2009, pp. 53-54). To uphold scientific rigor, the study adheres to the core methodological standards of objectivity, reliability and validity (Himme, 2009, p. 485). These criteria are essential for ensuring that the 18 findings are unbiased, reproducible and accurately capture the effects of the experimental manipulation. Objectivity is maintained through a standardized and transparent design that minimizes researcher influence. The survey wording, including all treatment conditions, is carefully crafted to avoid leading questions or emotionally charged language (see chapter 4.1.3.). A critical element is the use of random assignment of participants to different experimental conditions (Rack & Christophersen, 2009, pp. 19-20). Therefore, respondents are randomly assigned to one of four framing conditions: a cultural narrative frame, an EU frame or to one of the neutral control conditions. Randomization ensures that any observed differences in policy support can be attributed to the experimental manipulation rather than to pre-existing differences among participants (Mutz & Pemantle, 2015, p. 201). Furthermore, the survey uses a single-blind design, meaning participants are unaware of the existence of multiple versions of the policy text and do not know they are part of an experiment. This reduces the likelihood of demand effects or socially desirable responding, therefore preserving the integrity of the data (ibid, p. 196). Reliability, defined as the consistency and replicability of results under similar conditions, is supported through precise treatment definitions and consistent implementation. All participants assigned to a particular treatment condition receive identical stimuli, thereby ensuring internal consistency (Himme, 2009, p. 485). To further enhance reliability, the thesis avoids unnecessary complexity by limiting the number of control variables and keeping the questionnaire as concise and comprehensible as possible (Mutz & Pemantle, 2015, p. 210). Particular attention has been paid to designing the framing treatments in a way that maximizes their salience and distinctiveness (ibid, p. 194). Lastly, validity ensures that the thesis accurately measures what it claims to measure. Given that exposure to the experimental framing cannot be directly observed, a manipulation check will be included to verify that participants perceived the intended framing (Mutz & Pemantle, 2015, p. 196). This step is essential for confirming that the independent variable was successfully manipulated, an essential condition for establishing that an actual experiment has taken place (ibid, p. 193). To assess whether participants interpreted the framing as intended, a manipulation check will follow each treatment text (see chapter 4.1.3). This measure is crucial for ensuring internal validity, as it allows researchers to attribute observed differences in policy support to the experimental manipulation rather than to random variation or misinterpretation (Himme, 2009, p. 491). 19 4.1.3. Structure of the survey experiment Pilot study Before conducting the main experiment, a pilot study was carried out in March 2025 with 45 participants. The primary objective was to ensure that the pre-written treatment texts and survey questions were clearly formulated and understandable (Weimann & Brosig-Koch, 2019, pp. 202-203). The pretest took place in Germany using a face-to-face survey format, in which participants were randomly approached in public spaces and assigned to one of the treatment groups (TG) or control groups (CG). At the end of the survey, respondents were asked whether the policy description was understandable and were invited to provide feedback. In addition, to assess the effectiveness of the single-blind design, participants were asked what they believed the survey was about. The results of the pilot study provided valuable insights into the clarity and effectiveness of the different treatment conditions. In the first treatment condition (TG1), which framed the policy in terms of Germany’s engineering tradition and the role of synthetic fuels, some participants found the concept of climate-neutral synthetic fuels too abstract. In the second treatment condition (TG2), which emphasized the role of the EU in implementing the policy, some participants reported that they had already formed their opinion after reading the general policy description and did not consider the EU framing in their decision. While TG1 showed the expected increase in support, TG2 and the CG showed similar results. This may be because the EU framing was not strongly considered. When asked whether they perceived the policy as coming from the EU, some participants changed their opinion to a more opposed stance. This feedback indicates that the EU emphasis in the treatment text was not strong enough to create a noticeable framing effect. As a result, adjustments were made in the final version to ensure that the EU’s role in the policy was highlighted more clearly. Feedback from the CG, which received only the neutral policy description, confirmed that this version was clear and straightforward. However, a few participants indicated difficulty remembering the policy details, suggesting the need for improved delivery. In response, the policy description and support question were printed and read aloud, allowing participants to follow along visually. To assess whether participants were aware of the experimental manipulation, they were asked to describe what they believed the study was about. None of the participants correctly identified that they had been exposed to different versions of the policy text. Instead, responses varied widely, with some assuming the survey aimed to measure support for alternative fuels across age groups, others believing it focused on how people respond to sustainability policies and a few interpreting it as an investigation into public support toward EU regulations. These 20 responses confirmed that the single-blind design was successful, as participants were unaware of the true purpose of the pilot study. Based on these findings, several refinements were made before launching the main experiment. In TG1, an example of synthetic fuels was added to clarify the concept, together with minor wording adjustments. In TG2, the EU emphasis was strengthened to make the framing more salient. Furthermore, the order was adjusted so that the EU framing is mentioned first, leaving the policy suggestion unchanged in terms of content. The adjustment made merely serves to improve readability. In addition to the open-ended manipulation check, which serves to assess whether participants consciously recognize the topic and nature of the experiment, a second closed-ended question was added to check whether the treatments actually activated the desired social identification with the in-group (or its threat from an out-group). These manipulation checks are essential for the internal validity of the experiment, as they provide the necessary theoretical control to causally interpret the effect of framing on public support (Mutz & Pemantle, 2015, p. 196). As a result of this adjustment, the experimental design was expanded to include two TGs and two CGs, each matched to one of the treatment conditions. CG1 received only the in-group identity related item, while CG2 received the EU threat related item. This structure avoids potential bias that could result from presenting both questions in a single control group and creates two parallel mini-experiments, each testing a specific framing effect. While this expansion increases the total number of groups from two to four and therefore requires a larger overall sample size, it strengthens the internal logic of the design and the interpretability of the results. More generally, demographic questions were confirmed to be an effective way of easing participants into the survey and minor refinements were made throughout the questionnaire to improve clarity. To avoid potential bias, none of the participants from the pilot study will be included in the main study. Face-to-face survey experiment The final experiment followed a between-subjects design with four experimental groups: two TGs and two corresponding CGs. Each participant was randomly assigned to one of the groups during the face-to-face data collection process (see chapter 4.1.4.). All participants completed the same general sequence of questions, with slight variations introduced through the framing manipulation, depending on group assignment. At the beginning of the survey, participants were informed about the voluntary nature, anonymity and the approximate duration of three minutes. To avoid any influence, no 21 introduction was provided. It was merely mentioned that the survey concerns mobility policies. Participation required verbal consent. The survey then began with a short set of demographic questions (see chapter 4.4.1.). Next, all participants were presented with the same core policy description, which served as the neutral base for the framing manipulation. The policy stated that from 2035, newly registered vehicles will no longer be allowed to produce CO₂-emissions. This measure should help to ensure that the transport sector is climate-neutral from 2050. Depending on the condition, participants were then exposed to an additional framing paragraph. Participants in TG1 received a frame aligned with Germany’s national automotive identity, highlighting engineering excellence and the role of synthetic fuels in maintaining technological leadership: To uphold Germany’s tradition as a leading engineering nation with a strong automotive industry, climate-neutral synthetic fuels suitable for conventional refueling are being considered as a key solution. Participants in TG2 received a frame that emphasized the EU’s role and authority, portraying the policy as a collective European mandate requiring compliance from all member states: The European Union has determined that all member states, including Germany, must contribute to achieving the climate protection targets. Control groups CG1 and CG2 received only the neutral policy description without any additional framing. The overall structure is summarized in Table 1 below: Table 1: Framing conditions and manipulation check applied to each group TG1 CG1 TG2 CG2 Policy description X X X X Additional framing Cultural narrative framing EU framing In-group identity item X X Out-group threat item X X Immediately after the policy description, all participants were asked the same closed-ended question measuring their support for the policy with a rated five-point Likert scale (see chapter 4.1.1.). The policy description and the additional framing as well as the five-point Likert scale were printed, read aloud and shown to each participant, allowing respondents to follow the text visually. After indicating their level of support, all participants answered two manipulation checks, presented in the same fixed order. First, the following open-ended question was asked: What do you think the survey was about? This item was designed to assess whether participants had 22 recognized the experimental framing or inferred the underlying theme of the survey. Second, a closed-ended identity-related item was used to evaluate whether the respective framing had activated the expected social identity processes (manipulation check). This item varied depending on group assignment. TG1 and CG1were asked: Do you think that the car is inextricable linked to German culture?, triggering the cultural narrative as a key element of social identity. TG2 and CG2 were asked: Do you think that political interventions by the EU threatens Germanys cultural ties with the car?, triggering defensive reactions to the EU climate policy. This item was rated using the same five-point Likert scale as the policy support question. Finally, participants were thanked for their time and informed that the survey was complete. 4.1.4. Sample and data collection The primary data for this thesis was collected through a face-to-face survey between April 9 and April 22, 2025. The survey was based on pre-developed questionnaires, each including a short introductory text, demographic questions, the policy suggestion and the specific questions related to the experimental treatment (see Appendices A-D for the original German version and G-H for the translated English version). During the survey process, the questionnaires were presented orally by the interviewer, and participants’ responses were directly documented on a digital device. Data collection took place at three gas stations strategically selected to ensure sample diversity. One in a rural area, one in an urban area and one located on the Autobahn A7, a major highway connecting different German Länder. This geographic variation was intended to enhance the heterogeneity of the sample and improve the external validity of the findings (McDermott, 2012, p. 34). Gas stations serve as a fitting location, since they are closely linked to car usage and fuel consumption, which makes the environmental policy context immediately salient to participants (Manderscheid, 2020, p. 45). Moreover, they represent natural points of interaction with personal vehicles, making them particularly suitable for a thesis on environmental regulation. Approaching individuals at these locations increases the likelihood of engaging respondents who are actively involved in car use and therefore have a personal stake in policies such as the planned ban on ICEVs. This thesis aimed to ensure a real-world approach and study the support of people who are actually affected by the policy (Bergfeld et al., 2023, p. 1303). Furthermore, car drivers represent a highly relevant and representative group for studying public support for EU environmental policies in Germany. As of January 1, 2025, approximately 54 million people in 23 Germany hold a valid driver’s license (Kraftfahrt-Bundesamt, n.d.). According to the German Federal Statistical Office, approximately 70.3 million people aged 18 or older lived in Germany at the end of 2023. This means that roughly 76.8% of the adult population are licensed drivers (Statistisches Bundesamt, 2024). Given their size, it is also reasonable to assume that car drivers constitute a politically relevant electoral group, making them particularly important for research on support for EU environmental policies. In terms of daily mobility, cars account for 85% of motorized passenger transport in Germany, far surpassing other modes of transport. Moreover, 68% of employed individuals commute to work by car, illustrating the automobile’s dominant role in everyday life (Statistisches Bundesamt, 2025). These figures further underline that car drivers are not only socially and economically relevant, but also central to any discussion on environmental policies and behavioral change in the mobility sector. As the gas station can be seen as a symbolic space of automobility, participants were approached primarily while refueling, but also if they were using related services such as the shop or car wash. This also allowed for a practical approach, as gas stations provide a neutral and accessible environment where people are more likely to accept short interruptions and may have a few minutes available, which suits the survey’s compact design. The recruitment approach was designed to maximize engagement while minimizing disruption. Participants were selected using a random sampling method. Every person who agreed to participate was randomly assigned to one of four experimental groups. Randomization was implemented using a rotating sequence: the first participant was assigned to TG1, the second to TG2, the third to CG1 and the fourth to CG2. This cycle was then repeated continuously throughout the data collection process to ensure even group distribution (Mutz & Permantle, 2021, p. 201). The target population for this thesis comprises residents of Germany aged 18 and above who hold a valid driver’s license, as they are most directly affected by the policy under investigation. An a priori power analysis was conducted using G*Power 3.1 to determine the required sample size. Assuming a medium effect size (Cohen’s d = 0.5), the significance level (α) was set to 0.05 and statistical power (1–β) to 0.80, indicating an 80% probability of detecting a true effect. Based on these parameters, the analysis indicated a required minimum of 64 participants per group, resulting in a total of 256 participants overall (two treatments and two control). To account for possible non-responses/ non-participants or incomplete data, slight oversampling was planned. 24 Urban area The first station was located in a densely populated urban area in Hamburg, where car volume was significantly higher, which allowed for faster participant turnover. Data collection initially took place on April 11, 2025, in the morning hours. However, due to the weekday rush hour, many individuals declined participation citing stress and time constraints related to commuting. Within a short period, ten refusals were recorded, marking the highest rejection rate across all three locations. As a result, data collection was temporarily paused and resumed later the same day, at 4:00 p.m., when traffic was still high, but the atmosphere was more relaxed due to the end of the workday. Data collection at this location was completed on April 13, 2025. Despite the initial reluctance, many respondents showed interest in the topic once the survey structure was explained, particularly when they learned that the survey would take only a few minutes. A unique feature of this urban station was the presence of EV charging hubs, which allowed for the inclusion of more EVs users in the sample and added a modern mobility dimension to the otherwise fuel-oriented setting. Highway The second location was a highway service station on the A7, one of Germany’s major north– south transit routes. Data collection was conducted on April 17 and 18, 2025, with the second day added intentionally to coincide with a national public holiday and the start of a long weekend. Additionally, school holidays were ongoing in 15 of 16 federal states, increasing the probability of reaching travelers from different Länder (Schulferien, 2025). People were mostly approached while refueling or during short breaks and many were more willing to participate, as highway stations are generally designed for extended stops. Consequently, participants tended to have more time available than at other stations. An interesting observation was the higher proportion of respondents without private car ownership. In most cases, as participants themselves explained, this was due to the use of company vehicles. One methodological consideration at this location was the high presence of truck drivers. However, only private car drivers were approached for participation. This decision was made deliberately, as the focus of the thesis lies in personal support toward environmental mobility policies, rather than professional or commercial transport perspectives. By limiting the sample to private drivers, the thesis ensures consistency with its theoretical framework, which centers on personal identity (see chapter 3). 25 Rual area The third location was a rural gas station in Lower Saxony, which primarily served local residents. Data collection took place on April 19 and 20, 2025, during a vacation weekend, which resulted in a relaxed atmosphere. The lower population density meant that recruitment progressed more slowly, with longer pauses between interviews, but response willingness was very high. Only four individuals declined to participate, all citing disinterest rather than time constraints. In total, 264 individuals participated in the survey experiment. An additional 23 individuals were approached but declined to participate, citing various personal reasons or lack of interest. These cases are not included in the final sample or response rate calculation, as these individuals terminated the interaction before the start of the survey and thus were never exposed to the experimental material. Due to the short duration of the survey (approx. three minutes), most people participated. All 264 participants met the inclusion criteria, as they were approached while arriving at the gas station by car. Therefore, no one had to be excluded post-hoc based on eligibility. To assign participants to experimental groups, a sequential randomization procedure was implemented. The rotation followed a fixed, repeating order (TG1 > TG2 > CG1 > CG2). This cycle was then repeated continuously throughout the data collection process. This method of block randomization guaranteed an equal distribution of participants across all four groups, resulting in exactly 66 participants per group and 88 participants per area. 4.2. Limitations Despite careful design, several limitations must be acknowledged that may affect the interpretation and generalizability of this thesis findings. First, in terms of internal validity, one potential issue is the salience of the framing manipulations. Although the texts were carefully piloted and revised, it remains uncertain whether all participants consciously processed the intended narrative cues. This may have attenuated the effect of the treatments (Mutz & Permantle, 2015). Additionally, single-item measures are vulnerable to misinterpretation or random response error and do not allow for assessing the multidimensional nature of support (Greving, 2009, pp. 74-75). Including additional items could have strengthened the reliability of the dependent variable and helped confirm internal consistency in support levels. Second, external validity refers to the extent to which findings can be generalized beyond the specific context of the thesis (Rack & Christophersen, 2009, p. 27). This thesis employed 26 several strategies to enhance generalizability, including diverse gas station locations (urban, rural and highway) and random assignment. However, the scope of the thesis was limited to vehicle users in Germany. This excludes non-drivers and other relevant segments of the population. Furthermore, the findings may not generalize to other countries or to the broader German population, as the sample may not fully reflect all car owners’ attitudes. In addition, as noted in the literature review (see chapter 2.1.), previous research has shown significant regional differences in support toward environmental policies. While this thesis aimed to include some regional variation by selecting stations in different types of areas, practical constraints of a master’s thesis limited the data collection to Northern Germany, specifically, to Hamburg and Lower Saxony. A more geographically diverse sample would be necessary to make broader claims about national-level trends. Therefore, some bias needs to be acknowledged as not every car driver was able to be selected (Weimann & Brosig-Koch, 2019, p. 238). Third, although the experimental design was guided by SIT and incorporated a clearly defined cultural narrative, the operationalization of identity remains indirect. The thesis assumed, based on theory and the pilot study’s feedback, that the treatment frames activate identity salience. However, the complexity of identity processes likely exceeds what can be captured in a brief survey. 5. Data analysis and results5 In the following chapter, the statistical analysis conducted to test the thesis’ hypotheses is presented, followed by a detailed review of the results. Given the thesis structure as two parallel survey experiments, two separate Mann-Whitney U tests were conducted to assess the treatment effects independently. The dependent variable (public support) is measured using a five-point Likert scale. Likert-type items are best understood as ordinal data because while they indicate a ranked order, they represent ranked categories without guaranteeing equal intervals between response options (Greving, 2009, p. 73). For example, the psychological distance between neutral and somewhat agree may differ from that between somewhat agree and strongly agree. As a result, standard parametric tests such as the t-test, which assume interval-level measurement and normally distributed data, are not appropriate for analyzing this type of data (Walter & Rack, 2009, p. 392). Instead, the Mann-Whitney U test is a robust non-parametric alternative. It compares the 5 All results for each test are provided in the Appendix. See Appendix I for TG1/CG1 and Appendix J for TG2/CG2. The statistical tests were performed using the program R. The raw data can be made available upon request. 27 distribution of ranks between two independent groups without assuming normality or equal spacing between scale points (Nachar, 2008, pp. 13, 20). This makes it particularly well-suited for ordinal survey data, especially when based on a single item as in this thesis. Although the t- test is often simpler to conduct and interpret, it would be statistically inappropriate here unless assumptions about normality and interval scaling were met, which they are not (Walter & Rack, 2009, p. 392). Therefore, the Mann-Whitney U test helps preserve the integrity of statistical inference and ensures valid results under the actual conditions of the data (Nachar, 2008, p. 19). To confirm the need for non-parametric analysis, Shapiro-Wilk tests were performed to assess normality. All four groups showed significant deviations from a normal distribution: TG1 (W = 0.84278, p < 0.05), TG2 (W = 0.86574, p < 0.05), CG1 (W = 0.86293, p < 0.05) and CG2 (W = 0.85076, p < 0.05). These results justify the use of non-parametric testing. For each mini-experiment, a manipulation check was conducted to assess whether participants perceived the intended experimental framing. Participants answered a manipulation check question using the same five-point Likert scale. Again, Shapiro-Wilk tests revealed significant deviations from normality (p < 0.05), confirming the appropriateness of using Mann-Whitney U tests for the manipulation check as well. As previously mentioned, the thesis is organized as two parallel mini-experiments, each testing a different frame against its own control. Since the two TGs test distinct theoretical mechanisms and are not meant to be directly compared with one another, the appropriate analytical approach is to assess each treatment effect separately, rather than conducting a single omnibus test. 5.1. Cultural narrative frame The results for TG1, which was exposed to the cultural narrative frame, show a substantial and statistically significant effect on public support. A Mann-Whitney U test revealed a highly significant difference in public support between TG1 and CG1 (U = 670.5, p < 0.001), accompanied by a large effect size (r = 0.612). This indicates that the cultural framing had a strong influence on participants’ willingness to support the policy. Descriptive statistics support this finding, as shown in table 2 below. Table 2: Descriptive statistic, TG1 vs CG1 Treatment Mean Median Standard deviation (SD) TG1 (n = 66) 4.02 4 0.089 CG1 (n = 66) 2.41 2 1.16 28 Participants in TG1 reported an average policy support score of 4.02 (SD = 0.089), compared to a much lower mean of 2.41 (SD = 1.16) in CG1. The median values reinforce this contrast, with TG1 centered at 4 (agree somewhat) and CG1 at 2 (rather disagree). Moreover, the low standard deviation in TG1 suggests that responses were tightly clustered around higher support levels, indicating both higher and more uniform approval. Figure 1: Boxplot, TG1 vs CG1 The boxplot visualization (Figure 1) provides additional evidence of this effect. While CG1 exhibited a broader spread of responses, including an outlier at the higher end, TG1 displayed a tighter, more uniform distribution concentrated around strong support levels. This suggests that the cultural narrative did not merely shift the average opinion upwards, but also fostered a more homogeneous, collective endorsement among participants. In addition to measuring policy support, a manipulation check was conducted to assess whether participants in TG1 perceived the intended cultural narrative. The Mann-Whitney U test comparing manipulation check scores between TG1 and CG1 revealed a highly significant difference (U = 1389.5, p < 0.001) with a medium effect size (r = 0.323). Therefore, the manipulation check confirms that the TG1 indeed perceived the experimental framing as intended, strengthening the validity of the observed treatment effect on policy support. Lastly, to further confirm robustness of the findings, a permutation test was conducted as an additional non-parametric check (Weimann & Brosig-Koch, 2019, p. 173; Holt & Sullivan, 2023, p. 775). Permutation tests are particularly valuable because they do not rely on any assumptions about the distribution of the data or equality of variances. Instead, they estimate the reference distribution empirically by repeatedly reshuffling group labels and recalculating the test statistic across a large number of random permutations (Holt & Sullivan, 2023, pp. 776- 777). The permutation test confirmed the original findings, showing that the difference in public 29 support between TG1 and CG1 remained highly significant (p < 0.001). This result provides strong additional evidence that the observed treatment effect is not an artifact of distributional assumptions or small-sample biases. Overall, the results for TG1 provide strong evidence that framing environmental policies within a culturally resonant national narrative significantly increases public support. Not only did TG1 participants report higher and more consistent support compared to CG1, but the manipulation check further confirmed that participants meaningfully perceived the intended framing. 5.2. EU frame For the second mini-experiment, the Mann-Whitney U test comparing TG2 and CG2 revealed no significant difference in public support for the climate policy (U = 2221, p = 0.840). The corresponding effect size was insignificant (r = 0.018), indicating that the EU treatment had no meaningful impact on participants willingness to support the policy. Descriptive statistics further illustrate the lack of an effect (table 3). Table 3: Descriptive statistic, TG2 vs CG2 Treatment Mean Median Standard deviation (SD) TG2 (n = 66) 2.21 2 0.985 CG2 (n = 66) 2.35 2 1.26 Participants in TG2 reported an average policy support score of 2.21 (SD = 0.985), whereas participants in CG2 reported a similar mean of 2.35 (SD = 1.26). Both groups showed median support levels of 2, suggesting that support toward the policy remained relatively low and unchanged across conditions. The boxplot visualization supports this interpretation in Figure 2 below. 30 Figure 2: Boxplot, TG1 vs CG1 Both TG2 and CG2 exhibited wide variability in responses and no discernible upward shift in support was observed in TG. In contrast to the strong and concentrated effects seen in TG1, the data distribution for TG2 closely resembled that of its CG indicating that the EU framing failed to shift public opinion in a systematic way. To assess whether participants perceived the intended EU-themed framing, a manipulation check was again conducted. The Mann-Whitney U test revealed a statistically significant but small difference between TG2 and CG2 (U = 1702, p = 0.022) with a small effect size (r = 0.199). However, the weak effect implies that the manipulation was not strong enough to elicit a meaningful change in support. To verify the stability of these findings, a permutation test was conducted. The permutation test confirmed the Mann-Whitney U test results with no significant difference between TG2 and CG2 (p = 0.8439). This additional analysis supports the conclusion that the EU framing did not significantly affect participants policy support and strengthens the interpretation that the null result is robust across different inferential techniques. Overall, the results for TG2 suggest that framing environmental policies through an EU-threat lens did not significantly decrease public support. Both the main analysis and the robustness check confirmed the absence of a treatment effect. While a slight perception of the EU frame was detected through the manipulation check, the framing was likely too weak to shift public support. This result mirrors the pattern observed in the pilot study. 31 5.3. Randomization To ensure the comparability of TG and CG in both mini-experiments, randomization checks were conducted. The objective was to verify whether the distribution of relevant control variables (see chapter 4.1.1.) differed significantly between groups. Initially, chi-square tests were performed, as they are the standard approach for comparing categorical variables between groups (Riesenhuber, 2009, p. 13). However, due to small frequencies in several contingency tables, the chi-square approximation could not be reliably applied. Consequently, Fisher’s exact tests were employed instead, as they provide accurate results when cell sizes are small and the assumptions of the chi-square test are violated (Weimann & Brosig-Koch, 2019, p. 24). The results of the randomization checks are summarized in Table 4 below. Table 4: Randomization check, Fischer’s exact test Variable TG1 vs CG1 TG2 vs CG2 Age 0.1482 0.7717 Gender 1 0.8587 Education 1 0.0066** Car ownership and type 0.1368 0.8173 In the first experiment (TG1 vs CG1), no significant differences were found for any of the tested variables, suggesting that randomization was successful. However, in the second experiment (TG2 vs CG2), a significant difference in the distribution of educational attainment was observed (p = 0.0066). Consequently, education was controlled for in the subsequent analyses to account for this potential bias. To determine if this imbalance could have affected the main findings, a reduced linear regression model was estimated. This model included both treatment assignment (TG2 vs CG2) and education level as predictors of policy support. Other variables were excluded to focus the analysis specifically on the identified source of imbalance. In the model, no school-leaving certificate served as the reference category for the education variable and CG2 served as the reference category for the treatment variable. Therefore, the reported coefficients represent the difference in predicted policy support relative to participants assigned to CG2 with no school- leaving certificate. The results are summarized in Table 5 below. 32 Table 5: Reduced linear regression analysis Predictor Estimate p-value TG2 - 0.2085 0.2844 General school leaving certificate - 0.0655 0.8844 Professional qualification - 0.8045 0.0644* Academic degree 0.0987 0.8266 The model explained 14.0% of the variance in policy support (adjusted R² = 11.3%) and was overall significant (F (4, 127) = 5.184, p < 0.001). This indicates that treatment assignment and education together significantly improved the prediction of policy support compared to a null model without predictors. Nevertheless, after controlling for education, the treatment effect remained non-significant (p = 0.284), suggesting that the imbalance in education did not bias the main finding. Interestingly, the model also revealed that participants with a professional qualification showed a marginal tendency toward lower policy support compared to those without a school-leaving certificate (p = 0.064). Notably, this effect was independent of treatment condition and therefore reflects a general pattern, not a treatment-specific effect. To summarize, the regression analysis confirmed that the imbalance in educational attainment between TG2 and CG2 did not significantly affect the main outcome. Treatment assignment had no significant effect on policy support after controlling for education. The marginally lower support among participants with professional qualifications appears to be a general trend across both groups. 6. Discussion This chapter discusses the key findings of the survey experiment in relation to the theoretical framework and existing literature. Furthermore, it elaborates on their theoretical and practical implications, reflects on methodological limitations and outlines directions for future research. This thesis aimed to explore how Germany’s cultural narrative as a car-state influences public support for European environmental policies and whether different framings can trigger this effect. Drawing on SIT, two key hypotheses were tested. First, whether framing the policy in line with Germany’s automotive identity increases support (H1a) and second, whether presenting the same policy as an EU mandate decreases support (H1b). A face-to-face survey experiment with 264 participants tested these hypotheses through two mini-experiments using a cultural narrative frame, an EU frame and two neutral control conditions. 33 The results provide strong support for H1a. Framing the ICEV ban in line with Germany’s automotive identity significantly increased public support. This effect was not only statistically significant (p < 0.001) but also practically meaningful, with a large effect size and a narrower distribution of responses around high agreement. This finding aligns well with SIT’s central claim that individuals seek to preserve and affirm their in-group identity. It also supports Kunze’s (2022) argument that automobility is not merely an economic sector but a symbolically charged cultural narrative in Germany. By framing the policy as an opportunity to uphold national identity through innovation in e-fuels, the intervention neutralized perceived identity threats and reframed the policy as identity-affirming. In contrast, H1b was not supported. The EU frame, which emphasized compliance with supranational mandates, did not significantly decrease public support compared to the CG. Several factors may explain this outcome. First, the EU frame may have been too abstract or weak to trigger a strong out-group threat. As Fairbrother (2022) and Drews & van den Bergh (2015) emphasized, framing must resonate with existing symbolic meanings to be effective. The EU frame may have lacked emotional salience or relevance, failing to trigger a strong sense of out-group threat. Second, EU authority and with it the EU mandate may now be sufficiently normalized in Germany to no longer provoke strong emotional responses. Over time, EU environmental governance has become a routine part of policymaking. While the EU can be interpreted as an out-group in regulatory conflicts, it also represents a community of shared values, particularly in environmental protection. The absence of a clear emotional conflict may explain why the EU frame failed to provoke defensive reactions. This interpretation is supported by Hobolt and de Vries (2016), who argue that citizens differentiate less between national and EU-level governance in areas of shared competence, diminishing the potency of EU frames as identity threats. Third and lastly, as Hoppe et al. (2023) observed, abstract or top-down framings often fail to generate public engagement. From a SIT perspective, the EU can be perceived as an out-group, but unless this perception is emotionally salient and meaningful to respondents, it may not trigger defensive reactions. In this case, the EU frame likely failed to evoke the necessary sense of identity conflict or hierarchy disruption required for out-group derogation mechanisms to operate. Overall, the findings highlight that the right framing can significantly shape public support for EU environmental policies, but only when it aligns with cultural narratives. Theoretically, this thesis contributes to research on public support for environmental policies by demonstrating that identity-based cultural narratives are influential, yet underexplored, determinants of support. It provides empirical evidence for the claim that national identity can 34 be strategically leveraged to present supranational environmental regulation in more acceptable terms. Cultural narratives, rather than being static barriers, can function as dynamic resources for policy communication. These findings also matter beyond the academic realm. For the EU, they suggest that uniform policy mandates may be interpreted differently across member states, depending on their cultural narratives and identity frames. Rather than treating national identity as a barrier, policymakers could consider how to design communication strategies that resonate with member states’ self-conceptions. In Germany’s case, references to engineering pride or e-fuels, as in this experiment, can increase public support without altering the core substance of EU policy. More broadly, this implies that communication is not simply a technical step in policy implementation, it is a strategic component of policy design. This insight is particularly relevant for the success of the EGD. Given the EU’s cultural diversity, a one-size-fits-all communication approach is unlikely to succeed. National narratives provide a valuable entry point for tailoring policy frames to resonate with different member states’ identities. However, several limitations constrain the interpretation and generalizability of the results. First, the operationalization of identity was indirect. While the manipulation checks showed significant effects, especially in TG1, identity salience was not measured using multi-item scales or deeper psychological constructs. Second, although the sample was geographically diverse within Northern Germany, it excluded other regions with potentially different attitudes, particularly Southern Germany, where automobility is historically embedded and Eastern Germany, where policy skepticism has been observed (Georg et al., 2025; Mögele & Rau, 2020; Bumann, 2021). Third, single-item measurements of support, while suitable for the concise format of this thesis, limited the ability to explore the multidimensional nature of public attitudes. Moreover, while the cultural framing produced a large and statistically robust effect, the null result in the EU frame may partly reflect weaknesses in the salience or credibility of the manipulation. Future research could build on these findings in several ways. Follow-up experiments should use more robust identity measures, including scales capturing emotional attachment, perceived continuity and symbolic threat. Combining survey experiments with qualitative interviews could yield richer insights into how individuals make sense of identity-relevant frames. To improve the external validity of the results and build on the findings of this thesis, future studies should include participants from other regions of Germany, particularly southern and eastern 35 regions, where political support may differ. Additionally, while Germany’s car-state narrative is particularly prominent, the mechanism identified here could apply to other national contexts in the EU. For example, Italy offers a compelling case with its renowned car brands such as Ferrari, Fiat, Lamborghini and Maserati. These brands are not only key economic players but also serve as cultural icons that symbolize Italian craftsmanship. The Motor Valley region in Emilia-Romagna highlights how industrial production and regional identity can merge in powerful ways. Beyond the cultural symbolism of automobiles, countries with particularly high levels of car dependency may reveal similar patterns of identity-based resistance to environmental regulation (Alberti & Giusti, 2012). In Luxembourg, for instance, private car use remains high despite the introduction of free nationwide public transport in 2020, suggesting that convenience, infrastructure and established routines outweigh financial incentives (Gillard et al., 2024). Sweden further illustrates the importance of regional differentiation, while urban areas such as Stockholm benefit from comprehensive public transport systems and progressive mobility policies, rural areas continue to rely heavily on private vehicles (Rosqvist & Hiselius, 2019). The presence of a nationally car brand like Volvo may also reinforce a culturally specific car narrative. These insights highlight promising avenues for future comparative research on how national car cultures and mobility structures shape public attitudes toward environmental policies across Europe. While this study has focused on Germany’s automotive identity, the EU is characterized by a remarkable diversity of cultural, economic and historical contexts that shape national attitudes. Other member states may possess distinct cultural narratives that serve either to reinforce or to challenge EU environmental policy initiatives. For example, Eastern European countries, by contrast, might engage with environmental policies through narratives linked to modernization or post-socialist transformation (Blokker, 2005). Future research can build on these differences to explore how culturally embedded values and symbols mediate public responses to supranational governance. In doing so, this line of inquiry would not only extend the analytical reach of the present study but also provide deeper insight into the challenges of implementing cohesive environmental policies across a union as diverse as the EU. This thesis makes a valuable contribution to the field of European Studies by highlighting how a national cultural narrative can shape public support for EU environmental policy. It demonstrates how cultural narratives can hinder or facilitate the legitimacy and acceptance of EU environmental policies in a culturally diverse union. 36 7. Conclusion This thesis examined how the cultural narrative of Germany as a car-state influences public support for European environmental policies and how different framings of these policies affect that support. The thesis was guided by two research questions: (1) Does the cultural narrative of Germany as a car-state influence public support for European environmental policies and (2) how does the framing of the narrative influence the support? The findings provide a clear answer to the first question. Germany’s car-state narrative significantly influences public support for European environmental policies. The results show that framing the EU’s ICEV ban in a way that aligns with Germany’s automotive identity can significantly increase public support. A total of 93.9% of participants either supported or remained neutral toward the policy suggestion. This affirms that national cultural narratives are powerful factors in shaping public support for European environmental measures. Regarding the second question, the study shows that framing influences public opinion only if it resonates positively with national identity. Identity-affirming frames increased support. The EU mandate frame, which was intended to trigger perceptions of external pressure, did not significantly reduce support. This suggests that out-group threat framings are less effective if they lack emotional salience or fail to address identity-relevant issues. Overall, the results highlight that cultural narratives act as latent frameworks that influence public support. However, it is through targeted framing that these narratives become salient. Consequently, successful climate communication in the EU context should aim to connect supranational goals with positive aspects of national identity and cultural traditions in order to foster broader public acceptance. 37 References ADAC (2024): Synthetische Kraftstoffe: Sind E-fuels die Zukunft der Mobilität? Available at: https://www.adac.de/verkehr/tanken-kraftstoff-antrieb/alternative- antriebe/synthetische-kraftstoffe/ [16.05.2025]. Alberti, F.G. and Giusti, J.D. (2012): Cultural heritage, tourism and regional competitiveness: The Motor Valley cluster. City Culture and Society, 3 (4), pp. 261–273. doi: https://doi.org/10.1016/j.ccs.2012.11.003. Anderson, B., Böhmelt, T. and Ward, H. (2017): Public opinion and environmental policy output: a cross-national analysis of energy policies in Europe. Environmental Research Letters, 12 (11). Arora-Jonsson, S. (2014): Forty years of gender research and environmental policy: Where do we stand? Women’s Studies International Forum, 47, pp. 295–308. doi: http://dx.doi.org/10.1016/j.wsif.2014.02.009. Barnett, J., Graham, S., Quinn, T., Adger, W.N. and Butler, C. (2021): Three ways social identity shapes climate change adaptation. Environmental Research Letters, 16 (12). doi: https://doi.org/10.1088/1748-9326/ac36f7. Baute, S. (2024): How to foster public support for European climate policies: Evidence from the German population. Policy Paper (17). doi: https://nbn- resolving.de/urn:nbn:de:bsz:352-2-1cid5bwztha4f3. Bergfeld, M., Obersteller, D., Nguyen, D. M., Anderson, J. E. and Nobis, C. (2023): Real- world insights into user acceptance of e-fuels for passenger vehicles in Germany. Transportation Research Procedia, 72, pp. 1303–1310. doi: https://doi.org/10.1016/j.trpro.2023.11.591. Birel, T., Breeman, G.E., van Buitenen, A. and Vijver, M.G. (2024): Defueling the Impasse: EU Political Discourse on E-fuels. Energy Policy, 187. Doi: https://doi.org/10.1016/j.enpol.2024.114022. 38 Bergquist, M., Nilsson, A., Harring, N. & Jagers, S. C. (2022): Meta-analyses of fifteen determinants of public opinion about climate change taxes and laws. Nature Climate Change, 12 (3), pp. 235-240. https://doi.org/10.1038/s41558-022-01297-6. Blokker, P. (2005): Post-Communist modernization, transition Studies, and diversity in Europe. European Journal of Social Theory. Sage Publications, pp. 503–525. https://doi.org/10.1177/1368431005059703. Bundesregierung (n.d.): EU-Umweltrat: Nur noch CO2-frei fahren. Availabe at: https://www.bundesregierung.de/breg-de/service/archiv/verbrennermotoren-2058450 [Accessed: 17.05.2025]. Bumann, S. (2021): What are the Determinants of Public Support for Climate Policies? A Review of the Empirical Literature. Review of Economics, 72 (3), pp. 213–228. doi: https://doi.org/10.1515/roe-2021-0046. Carbo Europe (2025): Hydrogen Fuel vs. Electric Cars – Which is the Future of Green Transportation. Available at: https://www.carboeurope.org/hydrogen-fuel-vs-electric- cars/ [Accessed: 15.05.2025]. Cornet, A., Heuss, R., Schaufuss, P. and Tschiesner, A. (2023): A road map for Europe’s automotive industry. McKinsey & Company. Available at: https://worldtradescanner.com/a- road-map-for-europes-automotive-industry.pdf [Accessed: 16.05.2025]. Dhage, D.T. (2024): A critical review of environmental policy and political science approaches to climate change. International Journal of Research and Review, 11 (8), pp. 129–138. doi: https://doi.org/10.52403/ijrr.20240815. Drews, S. and van den Bergh, J.C.J.M. (2015): What explains public support for climate policies? A review of empirical and experimental studies. Climate Policy, 16 (7), pp. 855–876. doi: https://doi.org/10.1080/14693062.2015.1058240. 39 Drexler, C. E., Verse, B., Hauslbauer, A., Lopez, J. and Haider, S. (2022): Framing the mobility transition: Public communication of industry, science, media, and politics in Germany. Energy, Sustainability and Society, 12 (50). doi: https://doi.org/10.1186/s13705-022-00374-0. Dubois, M. (2001): Collective Identity and Expressive Forms. International Encyclopedia of the Social & Behavioral Sciences, pp. 2212–221. Doi: https://www.sciencedirect.com/science/article/pii/B0080430767040948. European Commission (n.d.): Transport and the Green Deal. Available at: https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/european- green-deal/transport-and-green-deal_en [Accessed: 16.05.2025]. European Commission (2022): Zero emission vehicles: first ‘Fit for 55' deal will end the sale of new CO2 emitting cars in Europe by 2035. Available at: https://ec.europa.eu/commission/presscorner/detail/en/ip_22_6462 [Accessed: 16.05.2025]. European Council (2025): Climate change: what the EU is doing. Available at: https://www.consilium.europa.eu/en/policies/climate- change/?utm_source=chatgpt.com [Accessed: 17.05.2025]. European Parliament (2024a): CO2 emissions from cars: facts and figures (infographics). Available at: https://www.europarl.europa.eu/topics/en/article/20190313STO31218/co2-emissions- from-cars-facts-and-figures-infographics [Accessed: 16.05.2025]. European Parliament (2024b): Review of EU electric-vehicle strategy and impact of decision to ban combustion-engine vehicles by 2035. Available at: https://www.europarl.europa.eu/doceo/document/E-10-2024-002171_EN.html [Accessed: 16.05.2025]. 40 European Parliament (2024c): The crisis facing the EU's automotive industry. Available at: https://www.europarl.europa.eu/RegData/etudes/ATAG/2024/762419/EPRS_ATA(20 24)762419_EN.pdf [Accessed: 16.05.2025]. European Parliament (2024d): Environment policy: general principles and basic framework. Available at: https://www.europarl.europa.eu/factsheets/en/sheet/71/environment- policy-general-principles-and-basic-framework [Accessed: 15.05.2025]. Entman, R. M. (1993): Framing: toward clarification of a fractured paradigm. Journal of Communication, 43(4), pp. 51–58. doi: https://doi.org/10.1111/j.1460- 2466.1993.tb01304.x. Fairbrother, M., Arrhenius, G., Bykvist, K. and Campbell, T. (2021): Governing for Future Generations: How political trust shapes attitudes towards climate and debt policies,' Frontiers in Political Science, 3. doi: https://doi.org/10.3389/fpos.2021.656053. Fairbrother, M. (2022): Public opinion about climate policies: A review and call for more studies of what people want. PLOS Climate, 1 (5). doi: https://doi.org/10.1371/journal.pclm.0000030. Gillard, M., Kębłowski, W., Boussauw, K. and Van Acker, V. (2024): “I always say, it’s the icing on the cake”: the discursive production of fare-free public transport in Luxembourg. Urban Geography, pp. 1–23. doi: https://doi.org/10.1080/02723638.2024.2412918. Goerg, S., Pondorfer, A. and Stöhr, V. (2025) Public support for more ambitious climate policies: Empirical evidence from Germany. Ecological Economics, 231. doi: https://doi.org/10.1016/j.ecolecon.2025.108538. Greving, B. (2009): Messen und Skalieren von Sachverhalten. In Albers, S., Klapper, D., Konradt, U., Walter, A. and Wolf, J. (eds.): Methodik der empirischen Forschung. 3rd edn. Wiesbaden: Springer Fachmedien Wiesbaden, pp. 65–78. 41 Haas, T. (2020): Die Mobilitätswende als Auslöser einer tief greifenden Transformation des „ „Modell Deutschland“? Politische Vierteljahresschrift, 62(1), pp.149-170. doi: https://www.jstor.org/stable/48768414. Harring, N. and Jagers, S.C. (2017): Why do people accept environmental policies? The prospects of higher education and changes in norms, beliefs and policy preferences. Environmental Education Research, 24 (6), pp. 791–806. doi: https://doi.org/10.1080/13504622.2017.1343281. Heyen, D.A. and Wicki, M. (2024): Increasing Public Support for Climate Policy Proposals: A Research Agenda on Governable Acceptability Factors. Climate Policy, 24 (6), pp. 785–794. doi: https://doi.org/10.1080/14693062.2024.2330390. Himme, A. (2009): Conjoint-Analysen. In Albers, S., Klapper, D., Konradt, U., Walter, A. and Wolf, J. (eds.): Methodik der empirischen Forschung. 3rd edn. Wiesbaden: Springer Fachmedien Wiesbaden, pp. 283–298. Hobolt, S.B. and De Vries, C.E. (2016): Public support for European integration. Annual Review of Political Science (19) 1, pp. 413–432. doi: https://www.annualreviews.org/content/journals/10.1146/annurev-polisci-042214- 044157. Holt, C.A. and Sullivan, S.P. (2023): Permutation tests for experimental data. Experimental Economics, 26 (4), pp. 775–812. Doi: https://doi.org/10.1007/s10683-023-09799-6. Hoppe, J., Patt, A. and Tröndle, T. (2023): Public support for phasing out carbon-intensive technologies: the end of the road for conventional cars in Germany? Climate Policy, 23 (7), pp. 813–828. doi: https://doi.org/10.1080/14693062.2023.2211042. Hornsey, M.J. (2008): Social Identity Theory and Self-categorization Theory: A Historical review. Social and Personality Psychology Compass, 2 (1), pp. 204–222. doi: https://doi.org/10.1111/j.1751-9004.2007.00066.x. 42 Kaya, M. (2009): Verfahren der Datenerhebung. In Albers, S., Klapper, D., Konradt, U., Walter, A. and Wolf, J. (eds.): Methodik der empirischen Forschung. 3rd edn. Wiesbaden: Springer Fachmedien Wiesbaden, pp. 49–64. Klas, A., Clarke, E.J.R., Fielding, K., Mackay, M., Lohmann, S. and Ling, M. (2022): Investigating how economic and national identity loss messages impact climate change policy support. Climatic Change, 175 (17). doi: https://doi.org/10.1007/s10584-022- 03472-2. Kraftfahrt-Bundesamt (n.d.): Fahrerlaubniserteilungen. Available at: https://www.kba.de/DE/Statistik/Kraftfahrer/Fahrerlaubnisse/Fahrerlaubniserteilungen /2023/2023_fe_e_tabellen.html?nn=3510850&fromStatistic=3510850&yearFilter=202 3&fromStatistic=3510850&yearFilter=2023 [Accessed: 16.05.2025]. Kunze, Conrad (2022): Deutschland als Autobahn. Eine Kulturgeschichte von Männlichkeit, Moderne und Nationalismus. Public History – Angewandte Geschichte (12). Bielefeld: transcript-Verlag. Mangum, M. and Block, R. (2018): Social Identity Theory and Public Opinion towards Immigration. Social Science, 7 (3). doi: https://doi.org/10.3390/socsci7030041. Manderscheid, K. (2020): Antriebs-, Verkehrs- oder Mobilitätswende? Zur Elektrifizierung des Automobilitätsdispositivs. In Brunnengräber, A. and Haas, T. (eds.): Baustelle Elektromobilität: Sozialwissenschaftliche Perspektiven auf die Transformation der (Auto-) Mobilität. Bielefeld: transcript, pp. 37–68. McDermott, R. (2012): Internal and external validity. Cambridge University Press eBooks, pp. 27–40. Doi: https://doi.org/10.1017/cbo9780511921452.003. Mögele, Mi. and Rau, H. (2020): Cultivating the "car state": a culturally sensitive analysis of car-centric discourses and mobility cultures in Southern Germany. Sustainability: Science, Practice and Policy, 16 (1), pp. 15–28. doi: https://doi.org/10.1080/15487733.2020.1756188. 43 Mutz, D.C. and Pemantle, R. (2015): Standards for experimental research: Encouraging a better understanding of experimental methods. Journal of Experimental Political Science, 2, pp. 192–215. doi: https://doi.org/10.1017/XPS.2015.4. Müller-Funk, W. (2007) Die Kultur und ihre Narrative. Eine Einführung. Springer Vienna. 2nd edn. Vienna: Springer Vienna New York. Nachar, N. (2008): The Mann-Whitney U: A Test for Assessing Whether Two Independent Samples Come from the Same Distribution. Tutorials in quantitative Methods for Psychlogy, 4 (1), pp. 13-20. doi: https://www.tqmp.org/RegularArticles/vol04- 1/p013/p013.pdf. Nielsen, K.S., Cologna, V., Bauer, J.M., Berger, S., Brick, C., Dietz, T., Hahnel, U.J., Henn, L., Lange, F., Stern, P.C. and Wolske, K.S. (2024): Realizing the full potential of behavioural science for climate change mitigation. Nature Climate Change, 14 (4), pp. 322-330. doi: https://doi.org/10.1038/s41558-024-01951-1. Nilsson, M. and Weitz, N. (2019): Governing Trade-Offs and building coherence in Policy- Making for the 2030 agenda. Politics and Governance, 7 (4), pp. 254–263. doi: https://doi.org/10.17645/pag.v7i4.2229. Rack, O. and Christophersen, T. (2009): Experimente. In Albers, S., Klapper, D., Konradt, U., Walter, A. & Wolf, J. (eds.): Methodik der empirischen Forschung. 3rd edn. Wiesbaden: Springer Fachmedien Wiesbaden, pp. 17–32. Rosqvist, L.S. and Hiselius, L.W. (2018): Understanding high car use in relation to policy measures based on Swedish data. Case Studies on Transport Policy, 7 (1), pp. 28–36. https://doi.org/10.1016/j.cstp.2018.11.004. Riesenhuber, F. (2009): Großzahlige empirische Forschung. In Albers, S., Klapper, D., Konradt, U., Walter, A. and Wolf, J. (eds.): Methodik der empirischen Forschung. 3rd edn. Wiesbaden: Springer Fachmedien Wiesbaden, pp. 1–16. 44 Statistisches Bundesamt (2024): Bevölkerung nach Altersgruppen. Available at: https://www.destatis.de/DE/Themen/Gesellschaft- Umwelt/Bevoelkerung/Bevoelkerungsstand/Tabellen/bevoelkerung-altersgruppen- deutschland.html [Accessed: 16.05.2025]. Statistisches Bundesamt (2025): Straßenverkehr: Dominanz des Autos ungebrochen. Available at: https://www.destatis.de/Europa/DE/Thema/Verkehr/Auto.html [Accessed: 16.05.2025]. Schulferien (2025): Schulferien Deutschland 2025, 2026. Available at: https://www.schulferien.org/deutschland/ferien/#google_vignette [Accessed: 16.05.2025]. Tajfel, H., and Turner, J. C. (2004). The Social Identity Theory of Intergroup Behavior. In Jost, J.T. and Sidanius, J. (eds.): Political psychology: Key readings. Psychology Press, pp. 276–293. Trepte, S. and Loy, L.S. (2017): Social Identity Theory and Self‐Categorization Theory. The international encyclopedia of media effects, 63, pp. 1-13. doi: https://doi.org/10.1002/9781118783764.wbieme0088. Tröndle, T., Annaheim, J., Hoppe, J., Hanger-Kopp, S. and Patt, A. (2023): Public preferences for phasing-out fossil fuels in the german building and transport sectors. Environmental Research Communications, 5 (8). doi: https://iopscience.iop.org/article/10.1088/2515-7620/acec39/pdf. Van Wieringen, K. (2024): The future of European electric vehicles. European Parliamentary Research Service, report PE 762.873. Available at: https://doi.org/10.2861/1724777 [Accessed: 17.05.2025]. Walter, S.G. and Rack, O. (2009): Eine anwendungsbezogene Einführung in die Hierarchische Lineare Modellierung (HLM). In Albers, S., Klapper, D., Konradt, U., Walter, A. and Wolf, J. (eds.): Methodik der empirischen Forschung. 3rd edn. Wiesbaden: Springer Fachmedien Wiesbaden, pp. 381–396. 45 Weimann, J. and Brosig-Koch, J. (2019): Die Reproduzierbarkeit von Experimenten. Einführung in die experimentelle Wirtschaftsforschung. Berlin: Springer Gabler. 46 Appendix Appendix A Questionnaire, TG1, German Guten Tag! Hätten Sie kurz Zeit, an meiner Umfrage für meine Masterarbeit teilzunehmen? Die Umfrage dauert etwa 3 Minuten. [Wenn „ja“ → weiter] Vielen Dank für Ihre Teilnahme! Ihre Teilnahme ist freiwillig und anonym. Sie können die Befragung jederzeit abbrechen. Wenn Sie fortfahren, stimmen Sie der Teilnahme zu. [Pause für Zustimmung] Als erstes werde ich Ihnen ein paar allgemeine Fragen stellen. 1. Wie alt sind Sie? o 18-29 o 30-49 o 50-66 o 67 oder älter 2. Wie würden Sie ihr Geschlecht beschreiben? o Mann o Frau o Diverse o Keine Angaben 3. Welchen höchsten Bildungsabschluss haben Sie erreicht? o Keinen Schulabschluss o Schulabschluss o Beruflicher Abschluss o Akademischer Abschluss 4. Besitzen Sie ein eigenes Auto? o Ja o Nein 4.1. Falls ja, welche Art von Auto besitzen Sie? [Mehrfachnennung möglich] o Fahrzeug mit Verbrennungsmotor (Benziner, Diesel, Gas etc.) o Fahrzeug mit Elektromotor o Hybridfahrzeug Als Nächstes werde ich Ihnen eine kurze Beschreibung einer politischen Maßnahme vorlesen. Danach bitte ich Sie, Ihre Zustimmung auf einer Skala von 1 bis 5 anzugeben. „1“ bedeutet „Stimme überhaupt nicht zu“, „5“ bedeutet „Stimme voll und ganz zu“. [Mit dem Finger auf ausgedruckten Text und Skala zeigen] Ich fange an vorzulesen: Ab dem Jahr 2035 dürfen neu zugelassene Fahrzeuge keine CO₂-Emissionen mehr verursachen. Diese Maßnahme soll dazu beitragen, dass der Transportsektor ab 2050 klimaneutral ist. Um Deutschlands Tradition als führende Ingenieursnation mit einer starken Automobilindustrie aufrechtzuerhalten, werden klimaneutrale synthetische Kraftstoffe, die gewohnt betankt werden können, als zentrale Lösung in Betracht gezogen. 5. Was ist Ihre Meinung zu dieser Maßnahme? o Stimme überhaupt nicht zu (1) o Stimme eher nicht zu (2) o Neutral (3) o Stimme eher zu (4) o Stimme voll und ganz zu (5) 47 Abschließend stelle ich noch zwei weitere Fragen. 6. Worum ging es Ihrer Meinung nach in der Umfrage? (Offene Frage) Bei der nächsten Frage sollen Sie wieder anhand der Skala antworten. [Erneut auf ausgedruckte Skala zeigen] 7. Denken Sie, dass das Auto untrennbar mit deutscher Kultur verbunden ist? o Stimme überhaupt nicht zu (1) o Stimme eher nicht zu (2) o Neutral (3) o Stimme eher zu (4) o Stimme voll und ganz zu (5) Das wars! Vielen Dank für Ihre Teilnahme! Appendix B Questionnaire, TG2, German Guten Tag! Hätten Sie kurz Zeit, an meiner Umfrage für meine Masterarbeit teilzunehmen? Die Umfrage dauert etwa 3 Minuten. [Wenn „ja“ → weiter] Vielen Dank für Ihre Teilnahme! Ihre Teilnahme ist freiwillig und anonym. Sie können die Befragung jederzeit abbrechen. Wenn Sie fortfahren, stimmen Sie der Teilnahme zu. [Pause für Zustimmung] Als erstes werde ich Ihnen ein paar allgemeine Fragen stellen. 1. Wie alt sind Sie? o 18-29 o 30-49 o 50-66 o 67 oder älter 2. Wie würden Sie ihr Geschlecht beschreiben? o Mann o Frau o Diverse o Keine Angaben 3. Welchen höchsten Bildungsabschluss haben Sie erreicht? o Keinen Schulabschluss o Schulabschluss o Beruflicher Abschluss o Akademischer Abschluss 4. Besitzen Sie ein eigenes Auto? o Ja o Nein 4.1. Falls ja, welche Art von Auto besitzen Sie? [Mehrfachnennung möglich] o Fahrzeug mit Verbrennungsmotor (Benziner, Diesel, Gas etc.) o Fahrzeug mit Elektromotor o Hybridfahrzeug Als Nächstes werde ich Ihnen eine kurze Beschreibung einer politischen Maßnahme vorlesen. Danach bitte ich Sie, Ihre Zustimmung auf einer Skala von 1 bis 5 anzugeben. „1“ bedeutet „Stimme überhaupt nicht zu“, „5“ bedeutet „Stimme voll und ganz zu“. [Mit dem Finger auf ausgedruckten Text und Skala zeigen] Ich fange an vorzulesen: 48 Die Europäische Union hat bestimmt, dass alle Mitgliedsstaaten einschließlich Deutschland zur Erreichung der Klimaschutzziele beitragen müssen. Das bedeutet, dass Ab dem Jahr 2035 [dürfen] neu zugelassene Fahrzeuge keine CO₂-Emissionen mehr verursachen dürfen. Diese Maßnahme soll dazu beitragen, dass der Transportsektor ab 2050 klimaneutral ist. 5. Was ist Ihre Meinung zu dieser Maßnahme? o Stimme überhaupt nicht zu (1) o Stimme eher nicht zu (2) o Neutral (3) o Stimme eher zu (4) o Stimme voll und ganz zu (5) Abschließend stelle ich noch zwei weitere Fragen. 6. Worum ging es Ihrer Meinung nach in der Umfrage? (Offene Frage) Bei der nächsten Frage sollen Sie wieder anhand der Skala antworten. [Erneut auf ausgedruckte Skala zeigen] 7. Denken Sie, dass politische Eingriffe der EU die deutsche kulturelle Verbundenheit mit dem Auto gefährden? o Stimme überhaupt nicht zu (1) o Stimme eher nicht zu (2) o Neutral (3) o Stimme eher zu (4) o Stimme voll und ganz zu (5) Das wars! Vielen Dank für Ihre Teilnahme! Appendix C Questionnaire, CG1, German Guten Tag! Hätten Sie kurz Zeit, an meiner Umfrage für meine Masterarbeit teilzunehmen? Die Umfrage dauert etwa 3 Minuten. [Wenn „ja“ → weiter] Vielen Dank für Ihre Teilnahme! Ihre Teilnahme ist freiwillig und anonym. Sie können die Befragung jederzeit abbrechen. Wenn Sie fortfahren, stimmen Sie der Teilnahme zu. [Pause für Zustimmung] Als erstes werde ich Ihnen ein paar allgemeine Fragen stellen. 1. Wie alt sind Sie? o 18-29 o 30-49 o 50-66 o 67 oder älter 2. Wie würden Sie ihr Geschlecht beschreiben? o Mann o Frau o Diverse o Keine Angaben 3. Welchen höchsten Bildungsabschluss haben Sie erreicht? o Keinen Schulabschluss o Schulabschluss o Beruflicher Abschluss o Akademischer Abschluss 49 4. Besitzen Sie ein eigenes Auto? o Ja o Nein 4.1. Falls ja, welche Art von Auto besitzen Sie? [Mehrfachnennung möglich] o Fahrzeug mit Verbrennungsmotor (Benziner, Diesel, Gas etc.) o Fahrzeug mit Elektromotor o Hybridfahrzeug Als Nächstes werde ich Ihnen eine kurze Beschreibung eine politische Maßnahme vorlesen. Danach bitte ich Sie, Ihre Zustimmung auf einer Skala von 1 bis 5 anzugeben. „1“ bedeutet „Stimme überhaupt nicht zu“, „5“ bedeutet „Stimme voll und ganz zu“. [Mit dem Finger auf ausgedruckten Text und Skala zeigen] Ich fange an vorzulesen: Ab dem Jahr 2035 dürfen neu zugelassene Fahrzeuge keine CO₂-Emissionen mehr verursachen. Diese Maßnahme soll dazu beitragen, dass der Transportsektor ab 2050 klimaneutral ist. 5. Was ist Ihre Meinung zu dieser Maßnahme? o Stimme überhaupt nicht zu (1) o Stimme eher nicht zu (2) o Neutral (3) o Stimme eher zu (4) o Stimme voll und ganz zu (5) Abschließend stelle ich noch zwei weitere Fragen. 6. Worum ging es Ihrer Meinung nach in der Umfrage? (Offene Frage) Bei der nächsten Frage sollen Sie wieder anhand der Skala antworten. [Erneut auf ausgedruckte Skala zeigen] 7. Denken Sie, dass das Auto untrennbar mit deutscher Kultur verbunden ist? o Stimme überhaupt nicht zu (1) o Stimme eher nicht zu (2) o Neutral (3) o Stimme eher zu (4) o Stimme voll und ganz zu (5) Das wars! Vielen Dank für Ihre Teilnahme! Appendix D Questionnaire, CG2, German Guten Tag! Hätten Sie kurz Zeit, an meiner Umfrage für meine Masterarbeit teilzunehmen? Die Umfrage dauert etwa 3 Minuten. [Wenn „ja“ → weiter] Vielen Dank für Ihre Teilnahme! Ihre Teilnahme ist freiwillig und anonym. Sie können die Befragung jederzeit abbrechen. Wenn Sie fortfahren, stimmen Sie der Teilnahme zu. [Pause für Zustimmung] Als erstes werde ich Ihnen ein paar allgemeine Fragen stellen. 1. Wie alt sind Sie? o 18-29 o 30-49 o 50-66 o 67 oder älter 50 2. Wie würden Sie ihr Geschlecht beschreiben? o Mann o Frau o Diverse o Keine Angaben 3. Welchen höchsten Bildungsabschluss haben Sie erreicht? o Keinen Schulabschluss o Schulabschluss o Beruflicher Abschluss o Akademischer Abschluss 4. Besitzen Sie ein eigenes Auto? o Ja o Nein 4.1. Falls ja, welche Art von Auto besitzen Sie? [Mehrfachnennung möglich] o Fahrzeug mit Verbrennungsmotor (Benziner, Diesel, Gas etc.) o Fahrzeug mit Elektromotor o Hybridfahrzeug Als Nächstes werde ich Ihnen eine kurze Beschreibung einer politischen Maßnahme vorlesen. Danach bitte ich Sie, Ihre Zustimmung auf einer Skala von 1 bis 5 anzugeben. „1“ bedeutet „Stimme überhaupt nicht zu“, „5“ bedeutet „Stimme voll und ganz zu“. [Mit dem Finger auf ausgedruckten Text und Skala zeigen] Ich fange an vorzulesen: Ab dem Jahr 2035 dürfen neu zugelassene Fahrzeuge keine CO₂-Emissionen mehr verursachen. Diese Maßnahme soll dazu beitragen, dass der Transportsektor ab 2050 klimaneutral ist. 5. Was ist Ihre Meinung zu dieser Maßnahme? o Stimme überhaupt nicht zu (1) o Stimme eher nicht zu (2) o Neutral (3) o Stimme eher zu (4) o Stimme voll und ganz zu (5) Abschließend stelle ich noch zwei weitere Fragen. 6. Worum ging es Ihrer Meinung nach in der Umfrage? (Offene Frage) Bei der nächsten Frage sollen Sie wieder anhand der Skala antworten. [Erneut auf ausgedruckte Skala zeigen] 7. Denken Sie, dass politische Eingriffe der EU die deutsche kulturelle Verbundenheit mit dem Auto gefährden? o Stimme überhaupt nicht zu (1) o Stimme eher nicht zu (2) o Neutral (3) o Stimme eher zu (4) o Stimme voll und ganz zu (5) Das wars! Vielen Dank für Ihre Teilnahme! Appendix E Questionnaire, TG1, Englisch (translated from German) Hello! Do you have a moment to take part in my survey for my master thesis? The survey takes about 3 minutes. [If “yes” → continue] Thank you for your participation! 51 Your participation is voluntary and anonymous. You can cancel the survey at any time. If you continue, you agree to participate. [Pause for consent] First, I will ask you a few general questions. 1. How old are you? o 18-29 o 30-49 o 50-66 o 67 or older 2. What is your gender? o Male o Female o Diverse o No specified 3. What is your education level? o No school leaving certificate o General school leaving certificate o Professional qualification o Academic degree 4. Do you own a car? o Yes o No 4.1. If yes, which type? [multiple answers possible] o Vehicle with combustion engine (petrol, diesel, gas, etc.) o Electric vehicle o Hybrid vehicle Next, I will read you a short description of a policy. I will then ask you to indicate your agreement on a scale of 1 to 5. „1“ means „strongly disagree “, „5“ means „strongly agree“. [Point to the printed text and scale with your finger] I start reading aloud: From 2035, newly registered vehicles will no longer be allowed to produce CO₂ emissions. This measure should help to ensure that the transport sector is climate-neutral from 2050. To uphold Germany’s tradition as a leading engineering nation with a strong automotive industry, climate- neutral synthetic fuels suitable for conventional refueling are being considered as a key solution. 5. What is your opinion on this policy? o do not agree at all (1) o rather disagree (2) o neutral (3) o agree somewhat (4) o strongly agree (5) Finally, I have two more questions. 6. What do you think the survey was about? (open question) For the next question, please answer again using the scale. [Point to the printed scale again] 52 7. Do you think that the car is inextricable linked to German culture? o do not agree at all (1) o rather disagree (2) o neutral (3) o agree somewhat (4) o strongly agree (5) That’s it! Thank you very much for your participation! Appendix F Questionnaire, TG2, Englisch (translated from German) Hello! Do you have a moment to take part in my survey for my master thesis? The survey takes about 3 minutes. [If “yes” → continue] Thank you for your participation! Your participation is voluntary and anonymous. You can cancel the survey at any time. If you continue, you agree to participate. [Pause for consent] First, I will ask you a few general questions. 1. How old are you? o 18-29 o 30-49 o 50-66 o 67 or older 2. What is your gender? o Male o Female o Diverse o No specified 3. What is your education level? o No school leaving certificate o General school leaving certificate o Professional qualification o Academic degree 4. Do you own a car? o Yes o No 4.1. If yes, which type? [multiple answers possible] o Vehicle with combustion engine (petrol, diesel, gas, etc.) o Electric vehicle o Hybrid vehicle Next, I will read you a short description of a policy. I will then ask you to indicate your agreement on a scale of 1 to 5. „1“ means „strongly disagree “, „5“ means „strongly agree“. [Point to the printed text and scale with your finger] I start reading aloud: The European Union has determined that all member states, including Germany, must contribute to achieving the climate protection targets. This means that… From 2035, newly registered vehicles will no longer be allowed to produce CO₂ emissions. This measure should help to ensure that the transport sector is climate-neutral from 2050. 53 5. What is your opinion on this policy? o do not agree at all (1) o rather disagree (2) o neutral (3) o agree somewhat (4) o strongly agree (5) Finally, I have two more questions. 6. What do you think the survey was about? (open question) For the next question, please answer again using the scale. [Point to the printed scale again] 7. Do you think that political interventions by the EU threatens Germanys cultural ties with the car? o do not agree at all (1) o rather disagree (2) o neutral (3) o agree somewhat (4) o strongly agree (5) That’s it! Thank you very much for your participation! Appendix G Questionnaire, CG1, Englisch (translated from German) Hello! Do you have a moment to take part in my survey for my master thesis? The survey takes about 3 minutes. [If “yes” → continue] Thank you for your participation! Your participation is voluntary and anonymous. You can cancel the survey at any time. If you continue, you agree to participate. [Pause for consent] First, I will ask you a few general questions. 1. How old are you? o 18-29 o 30-49 o 50-66 o 67 or older 2. What is your gender? o Male o Female o Diverse o No specified 3. What is your education level? o No school leaving certificate o General school leaving certificate o Professional qualification o Academic degree 4. Do you own a car? o Yes o No 54 4.1. If yes, which type? [multiple answers possible] o Vehicle with combustion engine (petrol, diesel, gas, etc.) o Electric vehicle o Hybrid vehicle Next, I will read you a short description of a policy. I will then ask you to indicate your agreement on a scale of 1 to 5. „1“ means „strongly disagree “, „5“ means „strongly agree“. [Point to the printed text and scale with your finger] I start reading aloud: From 2035, newly registered vehicles will no longer be allowed to produce CO₂ emissions. This measure should help to ensure that the transport sector is climate-neutral from 2050. 5. What is your opinion on this policy? o do not agree at all (1) o rather disagree (2) o neutral (3) o agree somewhat (4) o strongly agree (5) Finally, I have two more questions. 6. What do you think the survey was about? (open question) For the next question, please answer again using the scale. [Point to the printed scale again] 7. Do you think that the car is inextricable linked to German culture? o do not agree at all (1) o rather disagree (2) o neutral (3) o agree somewhat (4) o strongly agree (5) That’s it! Thank you very much for your participation! Appendix H Questionnaire, CG2, Englisch (translated from German) Hello! Do you have a moment to take part in my survey for my master thesis? The survey takes about 3 minutes. [If “yes” → continue] Thank you for your participation! Your participation is voluntary and anonymous. You can cancel the survey at any time. If you continue, you agree to participate. [Pause for consent] First, I will ask you a few general questions. 8. How old are you? o 18-29 o 30-49 o 50-66 o 67 or older 9. What is your gender? o Male o Female o Diverse o No specified 55 10. What is your education level? o No school leaving certificate o General school leaving certificate o Professional qualification o Academic degree 11. Do you own a car? o Yes o No 11.1. If yes, which type? [multiple answers possible] o Vehicle with combustion engine (petrol, diesel, gas, etc.) o Electric vehicle o Hybrid vehicle Next, I will read you a short description of a policy. I will then ask you to indicate your agreement on a scale of 1 to 5. „1“ means „strongly disagree “, „5“ means „strongly agree“. [Point to the printed text and scale with your finger] I start reading aloud: From 2035, newly registered vehicles will no longer be allowed to produce CO₂ emissions. This measure should help to ensure that the transport sector is climate-neutral from 2050. 12. What is your opinion on this policy? o do not agree at all (1) o rather disagree (2) o neutral (3) o agree somewhat (4) o strongly agree (5) Finally, I have two more questions. 13. What do you think the survey was about? (open question) For the next question, please answer again using the scale. [Point to the printed scale again] 14. Do you think that political interventions by the EU threatens Germanys cultural ties with the car? o do not agree at all (1) o rather disagree (2) o neutral (3) o agree somewhat (4) o strongly agree (5) That’s it! Thank you very much for your participation! Appendix I Treatment group 1 vs Control group 1 Shapiro-Wilk test Treatment W p TG1 0.84278 7.319e-07 CG1 0.86293 311e-06 56 Mann-Whitney U test U p 670.5 2.084e-12*** + Effect size r = 0.6117712 Descriptive statistic Treatment n Mean Median SD TG1 66 4.02 4 1.16 CG1 66 2.41 2 0.0886 Permutation test Z p -7.031 1e-0.4*** Manipulation test Shapiro-Wilk test Treatment W p TG1 0.80667 7.041e-08 CG1 0.90684 0.0001158 Mann-Whitney U test U p 1389.5 0.0002042*** + Effect size r = 0.3232401 Descriptive statistic Treatment n Mean Median SD TG1 66 3.98 4 0.936 CG1 66 3.15 3 1.28 Appendix J Treatment group 2 vs Control group 2 Shapiro-Wilk test Treatment W p TG2 0.86574 3.84e-06 CG2 0.85076 1.281e-06 Mann-Whitney U test U p 2221 0.8402 + Effect size r = 0.01754502 57 Descriptive statistic Treatment n Mean Median SD TG2 66 2.21 2 0.985 CG2 66 2.35 2 1.26 Permutation test Z p 0,20395 0.8439 Manipulation test Shapiro-Wilk test Treatment W p TG2 0.83867 5.523e-07 CG2 0.90918 0.0001435 Mann-Whitney U test U p 1702 0.02246* + Effect size r = 0.19867 Descriptive statistic Treatment n Mean Median SD TG2 66 3.08 3 0.751 CG2 66 2.68 3 1.17 58