Large datasets not necessarily represent broader population

Large datasets can create false confidence, as online platform users do not necessarily represent the broader population and online behaviour is often shaped by platform algorithms, Dr. Christina Monzer remarked about a good practice to use online datasets, while speaking at a webinar on Monday.

In her presentation, Dr. Christina Monzer discussed the promises and challenges of Computational Social Science (CSS), according to a press release.

Dr. Monzer is a postdoctoral fellow at the Politics, Identities and Communication Lab at the Annenberg School for Communication, University of Pennsylvania.

The Department of Journalism and Mass Communication at American International University-Bangladesh (AIUB) recently organised the webinar titled "Messy Data, Big Questions: Social Science in the Computational Age". Dr. Monzer attended the programme as the keynote speaker.

Assistant Professor Nasrin Akter hosted the programme. The session began with a welcome speech delivered by Head of the Department of Journalism and Mass Communication Prof. Dr. Pradip Kumar Panday, who emphasised the importance of integrating emerging research methods and interdisciplinary perspectives into journalism and communication education.

In his remarks, Pro-Vice Chancellor Prof. Dr. Abdur Rahman emphasised that in the age of unprecedented information flow, misinformation, disinformation, and malinformation pose serious threats to the information ecosystem, and that mass communication professionals have a crucial role in addressing these challenges.

Prof. Dr. AJM Shafiul Alam Bhuiyan noted that introducing journalism and communication students to new ideas and approaches such as computational research is essential for their academic and professional development.

Dr. Monzer explained how researchers can leverage large-scale digital data-often repurposed from online platforms-to explore important social questions.