Automating deductive content analysis in social science: From theory to practice
Bush House North East Wing, Strand Campus, London
About the Workshop
Have you ever wished for a faster, cheaper, yet effective way to annotate your huge dataset with a set of your codes/labels? Look no further. Attend this beginner-friendly workshop designed to introduce the conceptual and practical basics of automated frame (or more broadly, content) analysis [1][2].
You will learn, through a hands-on approach, how to enable a computer to annotate a dataset with labels of your choice [3][4], learn about the technical fundamentals and approaches (like finetuning) that make this possible [5][6], evaluate the efficacy of computers for this task, and discuss the existing limitations of such an approach.
The workshop will help sharpen your tech skills and open the door to an emerging methodology that could make large datasets manageable, especially for qualitative researchers. The workshop structure is straightforward; we introduce 'transformer' models [7], hyperparameters [8], and evaluation approaches [9] and jump right into implementing with these concepts in practice through AI-assisted scripting in Python.
Event Format
9:30 - 10:30: Public Talk
The fundamentals of automating content analysis in social sciences: An intuitive introduction
10:45 - 13:30: For registered participants only
Practical implementation of the methodology.
Speaker
Dr Vihang Jumle is a research associate at the Institute of Communication and Media Studies, University of Bern. His work examines how the design features and algorithmic logics of digital platforms shape the conditions under which citizens engage with, contest, or withdraw from democratic processes, and how mediated information environments influence political identities, public opinion, and perceptions of political legitimacy. Vihang's work has appeared in New Media & Society, Politics & Governance, Nations & Nationalism, Global Policy, amongst others and various news media outlets.
References
[1] Entman, R. M. (1993). Framing: Toward Clarification of a Fractured Paradigm. Journal of Communication, 43(4), 51–58.
[2] Krippendorff, K. (1989). Content analysis. International Encyclopedia of Communication, (226), 403-407.
[3] Raza, S., Bamgbose, O., Chatrath, V., Ghuge, S., Sidyakin, Y., & Mohammed Muaad, A. Y. (2024). Unlocking Bias Detection: Leveraging Transformer-Based Models for Content Analysis. IEEE Transactions on Computational Social Systems, 11(5), 6422–6434.
[4] Jumle, V. (2026). Clash of the models: Comparing performance of BERT-based variants for generic news frame detection (arXiv:2603.26156). arXiv.
[5] Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019). BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding (arXiv:1810.04805). arXiv.
[6] Terechshenko, Z., Linder, F., Padmakumar, V., Liu, F., Nagler, J., Tucker, J. A., & Bonneau, R. (2020). A Comparison of Methods in Political Science Text Classification: Transfer Learning Language Models for Politics. SSRN Electronic Journal.
[7] Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017). Attention Is All You Need. Advances in Neural Information Processing Systems, 30.
[8] Goodfellow, I., Bengio, Y., Courville, A., & Bengio, Y. (2016). Deep learning. Cambridge: MIT press. [9] Powers, D. M. W. (2020). Evaluation: From precision, recall and F-measure to ROC, informedness, markedness and correlation (arXiv:2010.16061). ArXiv.
Search for another event