This NSF-funded infrastructure project addresses longstanding barriers to accessing complex, user-generated data by creating a tool that opens social media datasets, like the COPE-ID corpus, to a broader spectrum of researchers. Traditionally, social, behavioral, and economic (SBE) scientists without programming expertise have been excluded from such work due to the technical demands of data acquisition via APIs and data processing pipelines. The DataViz tool eliminates the need for coding or computational fluency, enabling qualitative and mixed-methods researchers to engage with big data efficiently and meaningfully. By removing the manual, time-consuming barriers to data retrieval, the tool makes social media data more equitable and usable, especially for those who lack access to technical collaborators or resources. In doing so, it advances NSF’s mission to expand open-access data and promote broader participation, supporting cross-disciplinary innovation and empowering more inclusive research across a wide array of academic domains.

This project responds directly to Andreotta et al.’s (2019) call to integrate computational methods with social science in social media analyses, as one cannot make sense of textual data without classifying and capturing the contextual nuances of content, which machines cannot. Following Andreotta et al.’s (2019) proposed framework, DataViz supports a four-phased methodological approach designed to break down disciplinary silos and advance collaborative, mixed-methods research. This approach involved utilizing a large-scale social media dataset (i.e., COPE-ID), applying data science techniques, such as topic modeling (e.g., LDA, NMF, BERTopic), the social media posts into meaningful thematic structures, drawing down the most relevant topic clusters for closer scrutiny, and lastly creating a space where researchers can apply qualitative frameworks to interpret nuanced patterns within the data. This hybrid framework allows for scalable exploration of complex human behavior while preserving the interpretive depth of SBE research. In doing so, the tool enables transformative interdisciplinary research and bridges persistent gaps between computer science and the social sciences, underscoring the infrastructure grant program’s goal in fostering collaboration between fields that traditionally operate in silos.

The development of the DataViz tool, supported by NSF’s Infrastructure program, equips researchers to explore the complex societal dimensions of pandemics and disasters through interdisciplinary, data-informed approaches. By leveraging the large-scale COVID-19 dataset COPE-ID, the tool has the capability to support new theories of human behavior, which is critical as institutions continue to navigate the long-term impacts of COVID-19. For example, scholars across various disciplines, including communication, public health, psychology, political science, emergency management, and others, can investigate questions related to health messaging, institutional trust, and public sentiment. DataViz also enables the analysis of online behavioral patterns, providing insights into how communities respond to risk and public health interventions. This analysis can help identify how different populations adopt or resist health measures, with variations across time, space, and institutional contexts. Ultimately, the DataViz tool fosters researchers’ abilities to provide insights that support more transparent communication strategies, evidence-based decision-making, and equitable policy development for future public health crises and disaster resilience efforts.