DSAI Cohorts

Causal Effect Estimation in Observational Field Studies of Thermal Comfort

April 1, 2025
by Ruiji Sun. We introduce and apply regression discontinuity to thermal comfort field studies, which are typically observational. The method utilizes policy thresholds in China, where the winter district heating policy is based on cities' geographical locations relative to the Huai River. Using the regression discontinuity method, we quantify the causal effects of the experiment treatment (district heating) on the physical indoor environments and subjective responses of building occupants. In contrast, using conventional correlational analysis, we demonstrate that the correlation between indoor operative temperature and thermal sensation votes does not accurately reflect the causal relationship between the two. This highlights the importance of causal inference methods in thermal comfort field studies and other observational studies in building science where the regression discontinuity method might apply.

The Evolving Landscape of Web Scraping on Social Media Platforms

March 11, 2025
by Nanqin Ying. As social media platforms enforce stricter policies against unauthorized data collection, businesses and researchers must adapt to new API-based access models. This shift limits large-scale web scraping, impacting industries reliant on social media insights. The transition to paid API access and stringent compliance measures raises concerns about accessibility, cost, and ethical data collection. This article explores the evolving regulatory landscape, the enforcement of API restrictions, and how organizations can legally and ethically navigate data access in a world where scraping is becoming increasingly difficult. Understanding these changes is crucial for staying compliant while maintaining valuable insights from social media data.

Digitizing Inclusion: FinTech’s Promise and Pitfalls in the Global South

April 22, 2025
by Victoria Hollingshead. FinTech promises to revolutionize financial inclusion by harnessing data science to reach populations historically excluded from formal financial systems. By analyzing digital footprints, mobile payments, and behavioral data, startups and financial institutions have the potential to improve customer-lender interactions and revolutionize screening and monitoring techniques. While FinTech shows promise of enabling financial access, it also raises critical questions: how do we implement financial inclusion without reproducing the structures of the past? And more rhetorically, can financiers be the arbiters of financial inclusion, if their intrinsic role is to stratify and exclude?

Sharing Just Enough: The Magic Behind Gaining Privacy while Preserving Utility

April 15, 2025
by Sohail Khan. Netflix knows what you like, but does it need to know your politics too? We often face a frustrating choice: share our data and be tracked, or protect our privacy and lose personalization. But what if there was a third option? This article begins by introducing the concept of the privacy-utility trade-off, then explores the methods behind strategic data distortion, a technique that lets you subtly tweak your data to block sensitive inferences (like political views) while still maintaining useful recommendations. Finally, it looks ahead and advocates for a future where users, not platforms, shape the rules, reclaiming control of their own privacy.