As students learn how to program, both their programming code and their understanding of it evolves over time. In this work, we present a general data-driven approach, named "Temporal-ASTNN" for modeling student learning progression in open-ended programming domains. Temporal-ASTNN combines a novel neural network model based on abstract…

To establish an evidence base for future computer-assisted language learning (CALL) design, CALL research needs to move away from CALL versus non-CALL comparisons, and focus on investigating the differential impact of individual coding elements, that is, specific features of a technology which might have an impact on learning (Pederson, 1987).…

In this paper, we describe our learning progressions approach to early algebra research that involves the coordination of a curricular framework, an instructional sequence, written assessments, and levels of sophistication describing the development of students' thinking. We focus in particular on what we have learning through this approach about…

A longitudinal study of students' developing understanding of decimal notation has been conducted by testing over 3000 students in Grades 4 to 10 up to 7 times. A pencil-and-paper test based on a carefully designed set of decimal comparison items enabled students' responses to be classified into 11 codes and tracked over time. The paper reports on…

External representations have a great impact on what and how students learn. One key manner in which environments can operate upon novices' knowledge is through helping them ground their problem solving in an understanding of the situation embodied by the problem. In this paper, students' difficulties in microeconomics problem solving were…