Ankit, thanks for sharing this. For me, the most interesting part of the project is the "fine-grained logging technologies." This approach is central to several EdLab projects (e.g., Math Maze, JARS, EdLab CAS). It is also the foundation for doing work that advances efficiencies in education research.

I am also excited about the thinking about how this particular projects connects to our Hewlett Grant on adaptive education. The gathering of data and the reporting of it for formative evaluation is an important part of adaptive intervention.

"Unsupervised Discovery of Student Learning Tactics"

tactics - a group of behavior and actions, behaviors that have similar consequences of learning

student actions - an attempt vs hint-request, fast vs slow

"An Analysis of the Differences in the Frequency of Students’ Disengagement in Urban, Rural, and Suburban High Schools"

Difference in student across classrooms - suburban schools, urban schools, rural schools.

schools were compared on the basis of following disengaged behavior-
1. off-task behavior - talking to other student about Avatar movie
2. gaming the system
3. carelessness - making error despite knowing a skill

"Examining Learner Control in a Structured Inquiry Cycle Using Process Mining"

STAR Legacy inquiry cycle -- problems with problem based learning - students are lost in the sauce, they dont know how to proceed.

self assessment feedback and response - learners are given multiple attempts with different kinds of feedback.

clarify the question -> critique the response -> recommend resources (loop)