https://thesis.eur.nl/col/7039/ List of Publications en https://thesis.eur.nl/pub/65184/ Mon, 25 Jul 2022 00:00:01 GMT <div>Carola Weijers</div> The construction industry is increasingly adopting digital practices to improve processes and efficiency. Although they lack technological development compared to other sectors, they are making efforts towards becoming a more innovative, efficient sector by implementing various technologies onto the construction sites. The ASHVIN project aims to take this one step further through the implementation of sensors across its demonstration sites. Its measurements can, with the help of IoT (‘Internet of Things’) and digital twin technology, help make the sector more productive and efficient, while reducing costs. However, these technologies also have the potential to impact workers’ privacy and freedom within the workplace. Therefore, it becomes important to understand how the various stakeholders within the project make sense of these sensors and technologies to ensure a safe but privacy friendly workplace. This research addressed these issues by holding qualitative, in-depth interviews with various stakeholders within ASHVIN, together with the analysis of meeting minutes consisting of important discussions and decision-making processes. A thematic analysis of the interviews and meeting minutes was conducted. It was found that individuals discuss technologies in their workplace as a positive asset, as well as seeing no inherent issues with it. Technologies were also seen as contributing to an overall promising future with technological innovation at the forefront. Furthermore, sensors and technologies were seen to have surveillant properties in terms of monitoring practices and risks, as well as safety aspects. How employees working among the technologies were (or were not) informed was discussed through a construction worker perspective, as well as an overall level of awareness. Lastly, different contextual values associated with the technologies implemented were discussed, such as privacy and security. Overall, positive attitudes towards technologies and sensors were found and less attention given to its risks and privacy and security implications. Furthermore, privacy is often seen as a ‘box to be ticked’, rather than seeing the benefits associated with it. More attention should be paid to this, as well as aspects related to safety, privacy, and security in general. https://thesis.eur.nl/pub/65149/ Mon, 25 Jul 2022 00:00:01 GMT <div>Casper Kerklaan</div> The introduction of the General Data Protection Regulation (GDPR) in the European Union has brought important changes in how personal information gets distributed and handled online, by individuals and organisations. Applying the theory of contextual integrity, holding that to approach data subjects’ privacy we ought to look towards the flow of information in specific contexts, this study tests whether the collection and use of different types and amounts of personal data, collected by distinct public, private and hybrid organisations, result in significant differences in privacy concern among Dutch citizen-consumers. Moreover, individual factors have been considered, given how organisational trust was expected to negatively affect privacy concern, with less trust leading to more individual worry. Individual privacy knowledge, the extent to which people know how to handle and secure their data, was thought to strengthen this effect. To test these expectations, this study made use of an already existing dataset, created from a survey that was filled out by 510 Dutch participants, 311 of whom were included in the analyses of this study. To determine the role of contextual integrity in public privacy concern, respondents' answers on approximately 6912 unique vignette combinations were used, portraying hypothetical scenarios in which individuals shared data to different organisations, asking to what extent they felt worried about the sharing of their personal information in each circumstance. These vignettes were analysed using repeated measures ANOVA, also known as one-way within-subjects ANOVA, allowing for the testing of contextual differences in privacy concern depending on the organisation collecting the data, the amount of data collected, and the type of data collected. Additionally, a hierarchical regression analysis was conducted to account for individual factors influencing privacy concern, using survey responses outside the scope of the vignette study. This analysis could explain some of the variation in organisational trust, privacy knowledge, and privacy concern among individuals. These factors have been controlled for using age, levels of education, and frequency of smartphone use. The findings revealed that organisational trust was negatively related to privacy concern, however, privacy knowledge did not strengthen this relationship as a moderating variable, though it was still found to be a separate significant negative predictor of privacy concern instead. Significant differences were found in privacy concern depending on the organisation that collected the data, though this finding was only limited to some organisations, and varied scarcely depending on the type or amount of data. This study has provided new insights regarding public privacy concern in a context where strict privacy policies, courtesy of the GDPR, dictate how organisations can and should handle the personal data of the individual. The increased knowledge that the EU’s GDPR has provided its citizen-consumers has shown to have changed people’s perceptions on their rights to privacy and their concerns over organisations’ collection of their personal data. https://thesis.eur.nl/pub/66283/ Wed, 01 Feb 2023 00:00:01 GMT <div>Cindy Oscar Moore</div> YouTube is the second-most used social media platform worldwide and its popularity is partly due to its refined recommender system. This system provides users with specifically tailored video suggestions and it is based on an algorithm that is responsible for over 70% of the daily watchtime on the platform. The best-watched genre on YouTube concerns entertainment videos and the objective of this research is to inquire about user perceptions of the recommender system and how this affects entertainment content consumption. The study at hand provides a literature review on recommender systems, algorithmic imaginary, user consumption behavior and the uses and gratifications theory in light of the YouTube recommendation algorithm. However, literature that tied together user perceptions of the recommendation algorithm to YouTube entertainment content consumption was still lacking. Therefore, the research question of this study is as follows: To what extent do perceptions of YouTube's recommendation algorithm shape user entertainment content consumption? More specifically, it was studied whether user perceptions of the recommendation algorithm affect perceived entertainment content diversity, watchtime, recommendation satisfaction and perceived user agency. Moreover, it was inquired about whether content, social, process and/or technology gratifications influence recommendation satisfaction. A quantitative approach was implemented to measure these concepts using a survey (N = 161). The data was gathered among adult YouTube users who consume entertainment content, whereafter the data was statistically analyzed using SPSS. The results demonstrated that positive perceptions of the recommender system do not influence perceived entertainment content diversity, nor affect watchtime and neither influence the perceived amount of agency over the user's entertainment content consumption. However, more positive perceptions of the recommendation algorithm were found to lead to higher recommendation satisfaction. Also, content and technology gratifications were positively related to recommendation satisfaction, whilst social and process gratifications were not related. There are three key takeaways that can be derived from this study. Firstly, the attitude that the user holds of the algorithm does not influence the amount of time users spend watching YouTube videos. So, regardless of whether a user has a negative perception of the recommendation algorithm, the user does not necessarily watch less YouTube which might be related to the privacy paradox theory. Secondly, an interesting outcome from this study that substantiates the algorithm appreciation theory is that positive perceptions regarding the recommender system lead to more recommendation satisfaction. Thirdly, users who experience convenience while using YouTube were also more satisfied with their entertainment recommendations. The second and third key takeaway justify the reasoning for digital media companies to optimize and refine their platform’s algorithm, user interface and user experience as much as possible since this leads to higher satisfaction and therefore possibly to higher platform usage. The findings of this research contribute to the conclusions of previous studies and they might also be relevant to digital media companies, governmental entities and social actors. https://thesis.eur.nl/pub/71615/ Tue, 01 Aug 2023 00:00:01 GMT <div>Letizia Stefani</div> This study investigates the development of the Smart Road Plan, implemented by Anas S.p.A in Italy. This initiative aims to integrate mobility and accessibility into the urban fabric through digital solutions integrated into physical infrastructures for safety reasons. The road network functions and management is built on synergic integration among processes by obtaining interoperability and seamless system integration between citizens and infrastructures that cooperatively and synergically produce public value. Consequently, citizens have become both consumers and producers of services and information by continuously sharing data to the infrastructures and stakeholders through personal devices and on–board systems obtained directly from vehicles. This urban Renaissance strives to enable long–term economic growth while increasingly impacting citizens’ quality of life. Smart Road Plan represents a promising scenario to increase national competitiveness and sustainable prosperity development. This attractive scenario is positively portrayed for its effectiveness and beneficial application to significant issues and challenges. However, the road network has become a surveillant assemblage of information hubs whose collective intelligence is dictated by actionable data flows to sustain urban functionality. The concept of smartness may overlook potential harms and risks. The right to privacy is particularly threatened by advanced predictive analytics and hyper–monitoring of citizens from their spatial and temporal context. Furthermore, potential technological vulnerabilities to cybersecurity threats and attacks pose critical concerns about safeguarding personal and sensitive information. Then, it raises ethical and social issues relevant to citizens’ privacy who are concerned about their security due to the ubiquitous and pervasive penetration of services and platforms. However, the current literature fails to address the privacy and security implications pertaining real–time data exchanges between citizens and infrastructures. The Italian government and the European Commission have been closely working on policies and strategies to regulate the connectivity among users, vehicles, objects, and infrastructures. Therefore, it becomes important to explore how the Italian and European regulatory frameworks integrate citizens’ privacy and security with the technological progress in the Smart Road Plan. Qualitative content analysis in a case–study research design was conducted, with a mix of theory–driven and data–driven methodologies for developing a systematic coding scheme. The regulatory frameworks (N = 14) explore the Italian and European strategies and policies from relevant stakeholders and organizations involved. It was found that the regulatory context is fragmented and uncoordinated, characterized by a plurality of normative and strategical instruments. There is a lack of in–depth investigation of related risks and concerns, although privacy and security represent the biggest challenges and major worries of citizens in smart Renaissance. However, technical and organizational measures and standards are introduced. https://thesis.eur.nl/pub/71566/ Tue, 01 Aug 2023 00:00:01 GMT <div>Karlijn Maria van der Plaat</div> Increasingly, sex educators have ventured online to educate and inform people about sex and sexuality. Yet, their work is being hampered by social media platforms that use algorithmic systems to remove and reduce the visibility of sex-related content and accounts. Hence, online sex educators try to understand and master the algorithms that govern their work, as this may determine their existence on the platform. This thesis adopts the concept of the algorithmic imaginary as outlined by Bucher (2017) to research how online sex educators experience and perceive algorithmic content regulation on Instagram. It sets a methodological precedent by using phenomenological in-depth interviews to study users’ experiences with algorithms. Indeed, it finds that online sex educators have many encounters with algorithmic outputs as their content and accounts are shadowbanned and removed, which has fueled their algorithmic imaginary. This thesis adds on to the research about the algorithmic imaginary. It reconfirms that strange or odd outputs of algorithms grasp the attention of users and lead them to evaluate their workings. Yet, this research expands the idea of that which can be considered a strange encounter. Algorithmic content regulation of itself can be considered a strange encounter when users, like sex educators, feel wrongfully categorized as harmful while their only intent is to help and educate people. These strange encounters are a powerful instigator for users to reconsider how algorithms are used to exert particular belief systems, and it makes them believe that Instagram does not understand nor appreciate sex education. Additionally, this research provides new insights into the phenomenological study of algorithms. Bucher (2017) proposed the phenomenological approach as a new way of studying algorithms through the experiences of users that are affected by them. This research finds that it is difficult to talk of a unified experience when approaching algorithms using phenomenology, as algorithms provide a personalized output for every user. Yet, it argues that the essence of the experience with algorithms is its dynamic and volatile character. The ever-changing and inconsistent nature of algorithms causes feelings of insecurity and frustration for online sex educators. Users can never pinpoint the exact nature of the algorithm, which makes it easy for platforms to undermine their claims about unfair and biased regulation practices. Herein resides the power of the algorithm, and with that the power imbalance between Instagram and its users. Lastly, this research provides a new understanding into how users understand and perceive algorithms. Online sex educators do not view algorithms as a neutral technical system that functions in isolation of its social climate, rather they regard algorithms as another tool in the hands of powerful institutions that repress their efforts to empower people in their sexualities. There are many reports of sex workers, LGBTQIA+, BIPOC, fat people, disabled people, and more that experience similar forms of algorithmic content regulation on social media platforms. Hence, unpacking these experiences helps us to better understand the circumstances that these content creators encounter and the way in which social media platforms influence the public debate. https://thesis.eur.nl/pub/71567/ Tue, 01 Aug 2023 00:00:01 GMT <div>Daria Gusak</div> Recently, personalization of content became integral to social media platforms, as it optimizes business operations, increases customer satisfaction, and maintains a competitive edge. For social media users, personalization is convenient as it presents content based on their interests. With the importance of social media in people’s everyday life, questions are raised about the impact of algorithms on one’s identity. Algorithm responsiveness describes the extent to which personalized content relates to one’s identity, where algorithms either support one’s identity or ignore it. While algorithms may prioritize biased content or suppress a part of one's identity, users can lack cognitive resources to determine authentic content. Since personalized content relates to one’s identity, it shapes one’s self-perception, whilst users mutually shape personalized content by engaging with it. Nevertheless, people are generally not aware of algorithms or the impact of user behavior on personalized content and do not protect their privacy. Additionally, individuals’ perceptions of algorithms are contradictory, as some personalized content is refused and some is accepted. Scholars argue that algorithm awareness can determine the perception of algorithms, as when users become aware of algorithms they either appreciate the benefits of personalization or feel threatened. With algorithm awareness, users become more concerned about surveillance and are more likely to avoid personalized content, yet some scholars report the contrary. Since there is no research analyzing the aforementioned and as there are existing contradictory findings, this study aims to uncover “to what extent does algorithm responsiveness explain the impact of algorithmic awareness on personalization concerns of Instagram users living in the Netherlands?”. In this research, algorithm responsiveness is a mediator between algorithm awareness and the dependent variables surveillance concern and personalized content avoidance, while time spent on Instagram and age are control variables. By conducting a survey experiment (N=107), the level of participants’ algorithm awareness is manipulated, where explanations of algorithms are provided to two groups. However, the level of algorithm awareness did not have significant effects on algorithm responsiveness. Mediation analysis revealed that 8 out of 10 hypotheses are rejected, where algorithm responsiveness is not a significant mediator. Based on manipulated and non-manipulated algorithm awareness, perceived algorithm insensitivity increased the likelihood of personalized content avoidance. It was found that younger Instagram users perceive more responsiveness and less insensitivity and that they are more likely to avoid personalized content than older Instagram users. Based on the non-manipulated algorithm awareness, algorithm awareness increased perceived algorithm responsiveness. Thus, media entities should ensure that especially young users perceive algorithm responsiveness. Additionally, future research should utilize different stimuli and the scale for algorithm awareness and ensure a large and representative sample. The insignificant findings can be useful for scholars and policymakers, who place high importance on the level of algorithm awareness but should analyze other factors like age. https://thesis.eur.nl/pub/71616/ Tue, 01 Aug 2023 00:00:01 GMT <div>Alberto Di Rosa</div> As smart cities use data and technology to improve citizens’ quality of life in various areas, they also raise ethical and social challenges when employing nudging techniques to subtly and non-coercively influence citizens’ behavior and choices; these practices may shape moral and societal concerns regarding domains such as accountability, paternalism, manipulation, and invasion of privacy. Therefore, it’s crucial to comprehend how citizens perceive and evaluate such practices in light of smart city initiatives and objectives. This study, unlike most other studies that focus on experts or policymakers, offers a novel perspective based on ordinary people’s point of view by examining how citizens may perceive and assess nudging practices in smart cities, and how such practices are influenced by smart governance structures that use digital data flows to improve public decision-making and service delivery. On the social media platform Reddit, such ordinary people’s discourse is examined using a qualitative thematic analysis approach; five subreddits and twenty-five posts were selected using a purposive sampling method as they were deemed pertinent and representative of the research topic. Since no preexisting coding scheme was used, the inductive approach implied that the themes were to be identified and categorized based on the trends that appeared in the gathered data; ultimately, following the data analysis phase of this study, five were the themes that emerged: types of nudges, public opinion, government policing, smart city initiatives and goals, implications and opportunities. These themes assisted in addressing the research questions that were posed and, in addition, the findings demonstrated how nudges cause citizens to express a range of complex and varied opinions and emotions based on their personal experiences, background knowledge, perceptions of advantages and disadvantages, and ethical standards. However, they ultimately suggested nudging practices to be created and put into use in a human-centric manner that respects citizens’ autonomy, privacy, consent, while balancing innovation with regulation, and while also adjusting to the local context and culture that is in line with the general objectives and values of smart city concepts. This study contributes to the existing literature by offering a novel viewpoint on how smart governance should be developed based on ordinary people’s discourses; it also provides recommendations and insights for smart city practitioners and developers who may want to use nudging techniques in a human-centric manner. The findings’ implications for future research on ethically nudging citizens towards safer, healthier, and more sustainable behaviors are discussed, and arguments are made for promoting the need of constantly taking into account and evaluating citizens’ opinions. https://thesis.eur.nl/pub/71505/ Tue, 01 Aug 2023 00:00:01 GMT <div>Simone Sprangers</div> These days short-video platform TikTok is one of the most popular social media. While some research has already been done about how specific users use TikTok and how they think about issues such as privacy, a research gap exists between adolescents and older users, with a focus on English-speaking and Scandinavian cultures. To close this gap, this research project looked into young adult TikTok users in the Netherlands, particularly those aged 21-30, through four research questions: (1) How do young adult users in the Netherlands use TikTok in their day-to-day lives? (2) How do they perceive and interpret TikTok’s platform workings? (3) How do they perceive privacy issues surrounding TikTok? And lastly, (4) Do they use specific practices to manage their privacy on TikTok, and if so, how do they do that precisely? To answer these questions, I have conducted twelve qualitative in-depth interviews with active TikTok users aged 21-30 in the Netherlands. Afterward, the transcript data were analyzed through qualitative thematic analysis. The main findings recall that this group of TikTok-using young adults portrays specific ways of usage, of which entertainment and fun is the most significant, alongside gathering knowledge and insights they can use for their benefit, and documenting experiences. When they consume videos, they do so primarily by scrolling through the app and, in some cases, actively searching for content. Furthermore, all participants portray significant levels of TikTok insights into its platform workings, though to varying degrees, ranging from user activity and information collection and usage to quite technical algorithmic processes. The fact that TikTok has an extensive amount of control over its content, and the platform’s lack of transparency regarding data collection and usage, are points of critique that trouble a significant number of users, alongside the fact that the platform is subject to several lawsuits. Moreover, all participants find privacy important and seem rather knowledgeable about ways to protect their privacy on TikTok. Mentioned tactics are denying access to their location, confusing the platform by going against their habits, looking up random content to cloak their actual interests and identity, and keeping a close eye on their followers to keep out strangers. However, not all of them actually use these insights and tactics, or only to some extent. As such, there exist a number of paradoxes, such as the infamous privacy paradox, between what these young adult TikTok users think regarding privacy and what they do to protect it. Forces that seem to enforce this paradox are sentiments of nonchalance, circular reasoning that all social media collect data and therefore it is ‘ok’ and that TikTok cannot possibly be doing anything wrong since they have to follow regulations and are under strict observation by institutions such as privacy watchdogs. https://thesis.eur.nl/pub/75079/ Wed, 10 Jan 2024 00:00:01 GMT <div>Riemer, Eduard</div> Artificial Intelligence (AI) is becoming increasingly integrated into our society with applications ranging from facial recognition to self-driving cars. The potential benefits of these systems are vast, as everything we have achieved as civilization stems from intelligence. Recognizing the potential of AI, it is crucial to explore how we can harness its advantages while avoiding drawbacks (Future of Life Institute, 2015). Ensuring that these systems function safely and reliably is of the utmost importance. One way to achieve this is through the emerging field of AI auditing (Mökander et al., 2023, p. 6). As the consensus on AI shifts towards acknowledging its socio-technical character, it becomes essential to develop strategies that address issues like interpretability and various other social, legal, and technological challenges associated with this technology. A practical solution is embracing a Trustworthy Artificial Intelligence (TAI) approach (Thiebes et al., 2020, p. 447). In this thesis, TAI is approached from an AI auditing perspective by examining how AI experts conceptualize ethical, legal, and socio-technical elements and how these affect the realization of TAI. The thesis concludes that concepts that form the basis of the trustworthiness approach to AI systems are not easily operationalized because of their complex interpretability and subjectivity. To understand these multifaceted concepts clearly, we must view them from their social and cultural context. https://thesis.eur.nl/pub/75080/ Wed, 10 Jan 2024 00:00:01 GMT <div>Schumacher, Donna</div> Dutch political party leaders are Minecraft players, rappers, or simply silly, unserious individuals. At least, that is how they are represented in satirical sketches, lip syncing parodies, or memes. TikTok has increasingly become a creative outlet for creators to engage with politics. The platform has become one of the largest social media platforms, firmly embedding itself in the lives of Dutch citizens. TikTok becomes a groundbreaking stage where political conversations not only happen but also get embellished with a sparkle of humor. Given TikTok’s sociocultural impact and its growing role in shaping political narratives, understanding how TikTok creators navigate the political landscape is crucial. Therefore, this thesis aims to answer the research question: ‘How do TikTok- creators use humor in videos referring to politics during the Dutch parliamentary elections of 2023?’ Given that the platform is relatively new, yet has rapidly become one of the primary means for youth political engagement, this research is among the initial studies in this area. This study investigates: (1) the combinations of humor types and styles primarily found in TikTok videos referring to politics during the Dutch 2023 parliamentary elections, including the meaning-making styles and modes of engagement employed in them; (2) the use of humor types and attitudes towards the message or topic in these videos; and (3) the use of visual, audio, and editing elements to convey humor. This is accomplished through a mixed-methods approach, involving both quantitative and qualitative content analysis of 127 videos collected between two months before and one month after election day, November 22, 2023. The analysis yields three main findings. Firstly, the predominant humor types in the sample are ‘exaggeration’ and ‘silliness.’ Secondly, 80.4% of all posts contain a referential message. Thirdly, political humor videos during the period of the Dutch 2023 elections consist of an almost even distribution of virals and memes, each with its own distinctive characteristics, that are enhanced by the use of audio, visual and editing elements. TikTok thus emerges as a platform where a broad spectrum of possibilities coexist: a blend of silly nonsense and a significant role in the media landscape as a space for substantive political discourse. https://thesis.eur.nl/pub/75082/ Wed, 10 Jan 2024 00:00:01 GMT <div>Camp, Maaike van de</div> Social media has become a place where every scroll and click can shape beliefs and ideologies. As global tensions persist, social media platforms have risen beyond mere information channels to become influential places shaping perceptions and increasing polarization. This research delves into the intricate dynamics of how digital narratives on platforms such as Facebook, Twitter, and Instagram can swape opinions and deepen divisions. While social media has emerged as a potentially powerful force in sharing and shaping public opinion and discourse, there is only a limited understanding of how these dynamics manifest in already polarizing contexts. This research investigates the role of social media platforms in influencing public opinion regarding the Israel-Palestine conflict, exploring how digital narratives and the rapid dissemination of information can sway perceptions and fuel polarization. The research seeks to answer the question: How does the use of social media platforms influence the formation of public opinion regarding the Israel-Palestine conflict? The study provides a comprehensive literature review focusing on framing theory, agenda-setting theory, selective exposure, and incidental exposure to understand how and when certain perspectives are seen on social media. By examining previous literary work on social media and polarization, the study further analyses this polarization with a focus on tolerance through social identity theory and attraction and repulsion theory. A quantitative approach was implemented using a survey conducted among Dutch-speaking citizens, both active and inactive on social media. The findings reveal that exposure to pro-Palestine content on social media significantly increases support for Palestine, influenced by factors like echo chambers and agenda-setting. In contrast, pro-Israel content had less impact on support. Generational differences are also studied, with younger generations showing greater exposure to pro-Palestine content, which aligns with a wider trend of increased global empathy among younger people. Additionally, the study found that higher social media usage is associated with more participation in activist activities and, contrary to previous literature, greater tolerance towards the opposing party. The key takeaway from this study is that while social media can amplify polarizing perspectives, it also promotes greater tolerance towards opposing views. This research enhances our understanding of social media's impact on public opinion regarding polarizing issues. https://thesis.eur.nl/pub/74829/ Wed, 10 Jan 2024 00:00:01 GMT <div>Serkouh, Hind</div> Many Moroccan women of the diaspora experience forms of family surveillance; family members keep track of their activities and associations. This study aims to analyze the intersection between gender and surveillance in an intimate context where cultural precepts and Islam play a central role. To do so, this thesis employs Constructivist Grounded Theory analysis of interviews with Moroccan women of the diaspora to shed light on an existing phenomenon and expand on its contextual nuances rather than creating a new framework. The researcher being a Moroccan woman of the diaspora herself allowed her to establish an unparalleled level of trust and honesty with the interviewees about their shared experiences. While the opportunity to intimately engage with the respondents’ experiences was worthwhile, ethical challenges arose around navigating personal trauma as well as the respondents’ concern of their families discovering their strategies to resist family surveillance. To navigate these implications, common strategies such pseudonymization were used. Despite contextual distinctions and different family compositions, preliminary findings delineate a certain universality of experiences. All participants acknowledged their gender to be an essential parameter in the monitoring they were subjected to by their families; monitoring was accompanied by specific expectations and rules that were often implicit or difficult to negotiate -- such as those concerning their bodies, their sexuality, freedom of movement, and external perception. Participants who indicated receiving a Moroccan-culture-centered parenting tend to develop ‘double lives’ in order to circumvent said monitoring and internalize self-surveillance practices to avoid conflict and emotional disappointment. Those who instead received Islamic parenting have noticed lesser scrutiny and greater opportunities for communication and discussion of dynamics of surveillance. All participants identified love and care as the driving sentiment behind the monitoring; similarly, they acknowledged the struggle of the diasporic experience and the challenges of reconciling cultures with somehow clashing values. Ultimately, most participants confirm that they comply with cultural and religious rules to show respect to their families although they may not agree with them. This study introduces a perspective of family surveillance in the context of the Moroccan diaspora in Europe, a topic mostly studied in the context of Western realities. The research offers novel insights into how young women experience interculturality and religious culture while negotiating online and offline monitoring in transnational family life. https://thesis.eur.nl/pub/75083/ Wed, 10 Jan 2024 00:00:01 GMT <div>Visser, Famke</div> This research investigates the privacy invasive data gathering techniques used within menstruation-tracking applications (MTA). As these applications become increasingly popular due to the digitalisation of bodily functions and the ease of administering medical self-care, concerns for privacy grow as well. Oftentimes these applications are not transparent about their technological practices and ‘blackbox’ information and specifics of use algorithmic models. Simultaneously they continue to prompt users to give away their data, which serves as the fuel for functionality of the MTA. In answering the research question ‘How does the user-experience of menstruation-tracking applications relate to privacy and datafication?’, Critical Discourse Analysis is used to investigate thoroughly and critically seven in-depth interviews. Prior to the analysis interviews were held with both users of MTAs as well as privacy experts. Within the interviews they were asked to think about their privacy experiences and the possibility of the implementation of Privacy by Design (PBD). It resulted in finding that improvement of transparency, visibility, data minimisation and respect for user privacy is wanted by users, which define some of the principles of PBD. Moreover, the implementation of such principles are increasingly important as this research found that increased datafication of the body has created the growing urge to control the body, yet has also resulted in more people becoming in-sync with the tracking technology rather than their body. Finally, as more people become in-sync with this technology, the continuation of usage of, for example MTAs, comes with a growing risk to loose privacy, especially as this is ofter ‘rewarded’ with increased functionality of the technology. Ultimately, this research wishes to showcase how by bridging the gap between the cost-benefit considerations regarding user- and privacy-experience of users, companies and privacy experts, an increase in privacy does not always have to mean a loss in functionality. Yet only as long as companies are willing to transform their perspective on privacy protective measures such as PBD from something obligatory into an asset and selling point, can users experience an increase in privacy without a decrease in functionality. https://thesis.eur.nl/pub/75084/ Wed, 10 Jan 2024 00:00:01 GMT <div>Vianen, Lotte</div> Artificial Intelligence (AI) is becoming increasingly important for businesses to improve their work, and make it more efficient and user-friendly. Using user data, AI systems can create personalised recommendations in many different contexts, enabling widespread adoption of AI. Numerous determinants can influence people’s perception of AI, and it is necessary to find out how and when they influence people’s perception to be better able to influence people’s perception of AI when needed. This study therefore tries to find out whether knowledge of AI privacy implementations and the context in which an AI system is used affect people’s perception of AI. Knowledge about AI privacy implementations can differ amongst people because there are many different experiences and interests people have had, which could potentially influence people’s perceptions of AI. Additionally, AI systems are used in many different contexts, such as film recommendations and medical diagnosis, with varying degrees of importance of personal data that is used by the AI systems. The variable context will therefore be studied as a potential moderating variable in the relationship between knowledge of AI privacy implementations and perception of AI. Together, this leads to the question of “to what extent knowledge of privacy implementations in artificial intelligence has an effect on people’s perception of artificial intelligence, and whether this differs between the importance of the context”. A 2x2 between-subjects experiment (N = 131) was done to answer the research question. The two-way ANOVA test showed no significance for all effects. Knowledge of AI privacy implementations and the context in which the AI systems are used are both not significant determinants of people’s perception of AI. However, the moderating relationship was very close to significance and a power analysis revealed low power for the main analysis, so the decision was made to perform supplementary analyses. Here, perception of AI was split into trust in AI and usefulness of AI. These analyses showed that context is a significant determinant of trust in AI, and the moderating relationship between knowledge and context is also significant for trust in AI. Even though the main results were not significant, the mean differences show that less knowledge in a lower-stakes Netflix context leads to the most positive perception of AI, whereas little knowledge in a higher-stakes medical context works the opposite and creates the most negatively perceived perception of AI. This study therefore provides relevant, new and insightful information on the determinants of perception of AI and can help businesses that are working with AI. https://thesis.eur.nl/pub/75074/ Wed, 10 Jan 2024 00:00:01 GMT <div>Akinyi, Angeline</div> This study examines the use of ChatGPT among university students, exploring its impacts on academic work and the broader implications for higher education. As artificial intelligence (AI) continues to transform traditional learning methods, understanding how students utilize ChatGPT becomes crucial for educators and institutions. The research employs a qualitative approach, using in-depth interviews with university students who frequently use ChatGPT for academic purposes. The study is guided by five research questions, investigating how students use ChatGPT, its potential challenges and perceived benefits, institutional support needed, and the influence of the Unified Theory of Acceptance and Use of Technology on ChatGPT adoption. The findings revealed several key themes: Academic research and writing, idea generation, guidance and support informed the use of ChatGPT. Impact on creativity and cognitive skills, laziness, stigma, impact on traditional forms of testing, limited and inaccurate information, and prompt challenges informed the challenges associated with ChatGPT use. Efficiency and time saving, personalized learning and feedback, and enhanced productivity informed the perceived benefits of ChatGPT. Awareness and sensitization, training, integration into academic curricula answered the institutional support needed. Finally, the five constructs of UTAUT framework including, performance expectancy, effort expectancy, social influence, facilitating conditions and hedonic motivation highlighted the influence of the theory on ChatGPT adoption. The research concludes that while ChatGPT offers significant benefits for academic work, it also presents challenges that require careful consideration and management by both students and educational institutions. This study also provides valuable recommendations for supporting students in effectively utilizing ChatGPT while maintaining academic integrity and fostering critical thinking skills essential for the future workforce. https://thesis.eur.nl/pub/75076/ Wed, 10 Jan 2024 00:00:01 GMT <div>Mantel, Kyle</div> A key concern that emerges from the prevalence of media bias is political polarization, in which politically motivated groups on opposing sides become radicalized away from moderate center positions or collaboration, fueling extremism and intolerance. However, mainstream news sources, notably in the United States, reflect on these political parties and opposites, with potential bias a potential consequence of biased media reporting on key controversies and subjects. By analyzing the similarities and differences in reporting styles used by two politically opposed news sources, comparisons can be made to conclude whether and how political polarization can emerge from biased media reporting. This thesis conducts a thematic analysis on a key case study, analyzing Fox News and CNN coverage of the TikTok CEO hearing of 2023, in order to answer how politically polarized news sources report on a shared subject. Fox News and CNN were identified as politically-opposed media rivals and were determined to be appropriate for the case study, with a collection of 50 articles, an equal number of Fox News and CNN articles, collected for the coding process. With a thematic analysis, key themes were established with a coding process using the collected articles, centered around three focal areas, being mental health of users, national security concerns, and media coverage of the CEO hearing. The main research question is ‘How do American Politically Polarized News Sources report on the TikTok platform?’ This thesis concludes that both Fox News and CNN, which lean politically to the right and left respectively, share similar themes and patterns in reporting and can share agreement on specific subjects. However, political positions, ideologies, and perspectives can and do lead to differences in reporting, as well as more subtle differences ranging from grammar and key terms to the priority reporting of specific subjects. With regards to the TikTok CEO hearings of March 2023, the risk of political polarization is not as serious as suggested, at least within this specific case study. https://thesis.eur.nl/pub/75077/ Wed, 10 Jan 2024 00:00:01 GMT <div>Bruijn, Bente de</div> TikTok has become one of today’s most popular social media platforms, especially among Gen Z. As a subcategory of TikTok content, FitTok content is the TikTok equivalent of Instagram’s fitspiration content. While previous research suggesting the detrimental effects of social media and fitspiration consumption on body image and mental health is focused mainly on Facebook and Instagram, the current study explored the effects of fitspiration content on TikTok. Hence, the research was structured around the following central research question: “How does regular exposure to FitTok’s idealized body and life standards affect young adults’ body image and mental health?” Central to the operationalization of this research question were Festinger’s (1954) theory of social comparison and Bourdieu’s (1984) theory of practice based on the concepts of habitus and capital – two frameworks on the impact of social influences on individuals’ behaviors. To explore the relationship between FitTok consumption, social comparison, habitus and capital, and body image and mental health, a mixed-method study was conducted, combining an online survey (N = 150) and semi-structured in-depth interviews (N = 8). The sample consisted of young adult FitTok consumers (aged 18-25). By employing a convergent research design, the data of both methods were collected simultaneously, analyzed separately, and ultimately interpreted together. While no mediation effect for negative social comparison on the relationship between FitTok consumption and body image and mental health was found, a statistical analysis of the survey data suggested the importance of negative social comparison as a control variable in the relationship under examination. The interview data also revealed a general recognition of negative social comparison caused by FitTok content. However, for most of the interviewees, these comparisons did not negatively affect their body image and/or mental health. Rather, the interviewees emphasized the presence of positive social comparisons, contributing to improved mood. The demographic factors of gender, age (habitus), and education level (capital) were not found to impact the effect of FitTok consumption on body image and mental health. As these findings indicate, the ideals portrayed in FitTok content spark feelings of both negative and positive social comparison, thereby affecting young adults’ body image and mental health to different extents. https://thesis.eur.nl/pub/75078/ Wed, 10 Jan 2024 00:00:01 GMT <div>Stroosma, Lotte</div> This thesis investigates how consumers perceive surveillance-based marketing strategies employed by airline companies, focusing on the relationship between consumer surveillance, loyalty programs, and marketing strategies. Currently, airlines and other businesses increasingly rely on consumer data to personalize marketing efforts and target individuals with tailored advertisements. However, concerns about privacy and data security have prompted scrutiny of these practices and raised questions about their ethical implications. The research problem addressed in this study is the growing use of surveillance-based marketing strategies by airline companies and the impact on consumer perceptions. The main research question guiding this study is: How do consumers perceive airlines using surveillance-based marketing strategies to target them? To address this question, a qualitative research approach is employed, consisting of a thematic analysis of data obtained from the websites of five airline companies' loyalty programs and from ten interviews with consumers. The analysis focused on open, axial, and selective coding to identify patterns and themes related to consumer surveillance, marketing strategies, and data privacy. The findings reveal that consumers are generally aware of the use of surveillance-based marketing by airline companies but they show varying levels of concern about privacy and data security. While some consumers accept targeted advertising as a trade-off for personalized services, others express skepticism and distrust towards companies' data practices. Factors influencing consumer perceptions include transparency, trust, perceived benefits, and ethical considerations. The study highlights the need for greater transparency and accountability in data collection and use, as well as the importance of ethical marketing practices in maintaining consumer trust. It also highlights the complex relationship between consumer perceptions, marketing strategies, and ethical considerations in the digital era. In conclusion, this research contributes to our understanding of how consumers perceive surveillance-based marketing strategies in the airline industry and provides insights for airline companies and policymakers to navigate the ethical and privacy implications of data-driven marketing practices. It emphasizes the importance of balancing personalized marketing efforts with respect for consumer privacy rights and calls for greater attention to ethical considerations in marketing strategies. https://thesis.eur.nl/pub/76500/ Fri, 10 Oct 2025 00:00:01 GMT <div>Anja Ellwood</div> In 2024, all eyes turned to South Korea as it became the first victim of an epidemic of AI-generated deepfakes that disproportionately targeted women and minors. This technological threat, now surfacing worldwide, involves the non-consensual creation and distribution of hyper-realistic imagery, with current approaches to regulating this technology highlighting a societal unpreparedness and disconnect in understanding its true impact. This thesis thus explores the intricate issue of gendered digital violence, exploring how different groups perceive these harms, and the unexpected dual role of minors as both victims and perpetrators within this. Grounded in a virtual feminist theoretical framework, this multi-method qualitative research utilises focus groups and critical discourse analysis, with findings uncovering a 'Digital Violation Discrepancy' suggesting that the lack of women's perspectives, in AI development and regulation shapes understandings of deepfake harms. This discrepancy stems from a compounded issue: AI tools, instilled with patriarchal biases, birth an exploitative harm rooted in consent violation, further exacerbated by the crime's sui generis anonymity affordance that disrupts traditional legal and judicial proceedings relying on traceable evidence. These complexities were seen to be less understood by the male perspective, mirroring a critical gap in regulation and development measures that reflects this underrepresentation of women in these spheres. Consequently, deepfake creation's primary consent violation remains inadequately addressed in regulation, reflecting news representation where platform accountability is lacking, and fabricated harms, such as AI-generated child sexual abuse material, are normalised due to a misunderstanding of digital native behaviours. To address this, the research advocates for a foundational paradigm shift that prioritises women's experiences across all phases of AI development, ethical deliberation, and regulatory oversight, spanning AI governance and development reform, strengthening societal and educational responses, and encouraging international cooperation to harmonise legal frameworks. Ultimately, this thesis stresses that achieving an authentically equitable and secure digital future, safe from the uniquely gendered harms of deepfakes, requires challenging existing power structures within technology, ensuring AI empowers rather than exploits, particularly for women and minors. https://thesis.eur.nl/pub/76428/ Fri, 10 Oct 2025 00:00:01 GMT <div>Lucas Holef</div> This thesis examines how the policies and media discourse of the French radical right mobilize the argument of freedom of expression in response to the European Union's attempts to regulate the digital space, particularly through the Digital Services Act. In a context marked by the fragmentation of the information space, the rise of populism, and post-truth dynamics, the study questions the discursive strategies that allow regulation to be presented as a form of ideological censorship. Based on a corpus of parliamentary discourse analyses and press articles, the analysis tends to use the critical discourse analysis methods developed by Teun A. van Dijk, articulating this approach with a number of theoretical contributions, in particular those of Michel Foucault, Judith Butler, and Wendy Brown. The results highlight a set of recurring processes: lexicalized alarmism, the moral reversal of democratic principles, and a populist grammar that pits "the people" against "Brussels technocracy." Far from being anecdotal, these discourses are experienced as efforts to destabilize an order or a project in which the defense of freedom of expression constitutes, in a context of public regulation of digital technology, a counteroffensive.