Over the last decades and especially during the so-called "migration crisis" in 2015, migrants, migration and migration policies have become polarising political issues that have influenced and been influenced by public opinion not only in Slovakia but also in other European countries. Previous research abroad shows that media-mediated and framed information related to migration, migrants and migration policies plays a key role in shaping public opinion. Nevertheless, there is no research in Slovakia to date that focuses on the influence of media discourse on migration attitudes. There are also no analytical tools in the Slovak language to date for exploration of media discourse based on a robust analysis of a large amount of media content.

The aim of our project is to implement an integrative research programme on the impact of media discourse on migration attitudes and to develop innovative analytical tools for its implementation. In the project, we will leverage the interdisciplinary potential of the research team, which will allow us to use innovative approaches to computational natural language processing combined with expertise in media discourse and migration attitudes research. Through automated content analysis and complementary qualitative and quantitative observational and experimental methods, we will explore the longitudinal dynamics of the relationship between real events, media discourse and migration attitudes; the impact of media representations of migration categories, gender categorical typicality in media representations of migrants, and media signal about migration policy on migration attitudes; and the impact of actual media consumption on knowledge about migration and migration attitudes. We will also develop methods, techniques, resources and tools for automated analysis of textual content focused on migration issues in the Slovak language, which will also be useful for exploring topics other than migration.

Project goals

1. Develop methods, techniques, resources and tools for the automated analysis of textual content focused on migration issues in the Slovak language, which will also be useful in the study of topics other than migration.
2. Investigate the long-term dynamics of the relationship between real-world events (e.g. the so-called "migration crisis" in 2015), media discourse (media coverage, framing, emotions and sentiments related to migration) and attitudes towards migration, migrants and migration policies in the period from 2003 to 2022. The research will be carried out through automated analysis of media content and analysis of published opinion polls in the period in question.
3. Examine the impact of media representations of migration categories (refugee(s), migrant(s), asylum seeker(s), and foreigner(s)) on attitudes towards migration, migrant people, and migration policies. The research will be conducted through automated content analysis, cognitive interviews and survey experiments.
4. Examine the impact of gender categorical typicality in media representations of migrant people (migrant people referred to as 'migrants' or as 'male and female migrants') on attitudes towards migration, migrant people and migration policies. The research will be conducted through automated content analysis, cognitive interviews and survey experiments.
5. Examine the impact of the media signal about migration policy (media coverage of the problems facing migration policies and the solutions they offer) on attitudes towards migration, migrants and migration policies. We will conduct the research through automated content analysis, cognitive and semi -structured interviews, and conjoint and survey experiments.
6. Examine the impact of actual media consumption (consumption of media content in traditional and online, public and commercial, broadsheet and tabloid media) on knowledge about migration and on attitudes towards migration, migrants and migration policies. We will conduct the research through automated content analysis, cognitive interviews, multimedia electronic measurement of media consumption, a panel survey, and conjoint and survey experiments.

The four-year long project has a budget of 250,000 euro. It is supported by the Slovak Research and Development Agency.

• Hamerlik, E., Šuppa, M., Blšták, M., Kubík, J., Takáč, M., Šimko, M., Findor, A.: ChatGPT as Your n-th Annotator: Experiments in Leveraging Large Language Models for Social Science Text Annotation in Slovak Language. Proceedings of the 4th Workshop on Computational Linguistics for the Political and Social Sciences (CPSS), 2024.
• Vlasceanu, M., Doell, K. C., Bak-Coleman, J. B., Todorova, B., ... Findor, A., … Van Bavel, J. J. (2024). Addressing climate change with behavioral science: A global intervention tournament in 63 countries. Science Advances, 10(6), eadj5778. https://doi.org/10.1126/sciadv.adj5778
• Doell, K. C., Todorova, B., Vlasceanu, M., Bak Coleman, J. B., Pronizius, E., Schumann, P., Azevedo, F., … Findor, A., … Van Bavel, J. J. (2024). The International Climate Psychology Collaboration: Climate change-related data collected from 63 countries. Scientific Data, 11(1), 1066. https://doi.org/10.1038/s41597-024-03865-1
• Acs J, Hamerlik E, Schwartz R, Smith NA, Kornai A. Morphosyntactic probing of multilingual BERT models. Natural Language Engineering. Published online 2023:1-40. doi:10.1017/S1351324923000190
• Šuba, D., Šuppa, M., Kubík, J., Hamerlik, E., Takáč, M.: WikiGoldSK: Annotated Dataset, Baselines and Few-Shot Learning Experiments for Slovak Named Entity Recognition. In Piskorski, J. et al. (eds.), Proceedings of the 9th Workshop on Slavic Natural Language Processing 2023 (SlavicNLP 2023), Association for Computational Linguistics. 138-145. 2023.
• Kubík, J., Kyselica, D., Takáč, M.: Efficient fine-tuning of SlovakBERT with Epinet. In Brejová, B. et al. (eds.), ITAT 2023 Information Technologies – Applications and Theory 2023, Aachen. 96-101. 2023.
• Hamerlik, E., Deb, M., Takáč, M.: Bi-Source Class Visualization: An Adversarial Neural Network-based Approach for Unbiased Class Visualization. In World Symposium on Digital Intelligence for Systems and Machines (DISA2023), New York : Institute of Electrical and Electronics Engineers, 78-83. 2023.

Theses

• Tomášiková, D.: Gender stereotypes and voter preferences in Slovakia. Masters thesis. FSEV UK. 2023. Supervisor: doc. PhDr. Andrej Findor, PhD.
• Gajdoš, O.: Analýza sentimentu v príspevkoch o utečencoch na sociálnych sieťach. Bachelor thesis. FMFI UK. 2023. Supervisor: Mgr. Jozef Kubík, Konsultant: doc. RNDr. Martin Takáč, PhD.
• Ranincová, B.: Exploring the Influence of Gender on Slovak Public Perceptions Towards People Fleeing the War in Ukraine. Bachelor thesis. FSEV UK. 2023. Supervisor: doc. PhDr. Andrej Findor, PhD.
• Farkaš, D.: Improving Self-reported Media Consumption Measurement through Semi-structured Interviews and Cognitive Interviews. Bachelor thesis. FSEV UK. 2023. Supervisor: doc. PhDr. Andrej Findor, PhD.
• Prejsa, N.: Vnímanie pojmu migrant/ka slovenskou verejnosťou. Bachelor thesis. FSEV UK. 2023. Supervisor: doc. PhDr. Andrej Findor, PhD.
• Dóša, R.: Vnímanie pojmu odídenec/ka slovenskou verejnosťou. Bachelor thesis. FSEV UK. 2023. Supervisor: doc. PhDr. Andrej Findor, PhD.

• Facebook of Department of applied informatics FMFI UK, 15. 10. 2024