[ENG]
Vast amounts of data are becoming increasingly available to capture social processes at large scale and in great detail. Meanwhile, the rapid development of AI and computational approaches offers powerful toolkits for analyzing various data. In this talk, I will use two examples to illustrate how large-scale data and computational tools can be leveraged to advance our understanding of higher education social systems. First, I will discuss how to infer individual behavioral patterns from millions of digital records. Specifically, we developed an orderliness measure for student behavioral regularity and found that it has predictive power for academic performance. Second, I will discuss how to quantify student interactions based on administrative data. Specifically, we exploited plausible random assignment of college roommates to measure peer effects and studied the impact of peer heterogeneity on academic performance.
[ESP]
Grandes volúmenes de datos están cada vez más disponibles para capturar procesos sociales a gran escala y con gran detalle. Mientras tanto, el rápido desarrollo de la inteligencia artificial y los enfoques computacionales ofrece potentes herramientas para analizar diversos datos. En esta charla, utilizaré dos ejemplos para ilustrar cómo los datos a gran escala y las herramientas computacionales pueden aprovecharse para avanzar en nuestra comprensión de los sistemas sociales en la educación superior. En primer lugar, discutiré cómo inferir patrones de comportamiento individual a partir de millones de registros digitales. Específicamente, desarrollamos una medida de orden para la regularidad del comportamiento estudiantil y encontramos que tiene poder predictivo para el rendimiento académico. En segundo lugar, analizaré cómo cuantificar las interacciones entre estudiantes basándonos en datos administrativos. Específicamente, explotamos la asignación aleatoria plausible de compañeros de cuarto en la universidad para medir los efectos de los pares y estudiamos el impacto de la heterogeneidad de los compañeros en el rendimiento académico. (Traducido por ChatGPT).
Dr. Jian Gao is a Research Assistant Professor at the Center for Science of Science and Innovation (CSSI), Kellogg School of Management, Northwestern University, USA. His research interests lie in the interdisciplinary fields of Computational Social Science and Science of Science. He uses large-scale datasets and develops computational approaches to study complex social and economic systems. His recent work focuses on AI for science, the use of science by social institutions, and the workforce that drives science and innovation. His research has been published in leading journals, such as Science, Physics Reports, and Nature Communications, and covered by global media outlets, such as Nature News, Scientific American, and Forbes.