According to Chi's interactive-constructive-active-passive (ICAP) framework (Chi 2009; Chi and Wylie 2014), there are four modes of cognitive engagement based on students' overt behaviors.
The purpose of this study, however, was not limited to merely explaining variation in success based on student behaviors, but to identify and explain mediating relationships among the components of success.
Another important limitation is that the conclusions of this study are based on students' perceptions and not on observed behavior.
However, as recognized by the authors, a limitation to both studies was that data were obtained based on students' perceptions and retrospective self-report rather than on their actual behaviors.
"Respect for students' abilities and experiences," "the role of instructor in organizing the content and sequence of learning based on students' needs," and "the instructor should measure teaching effectiveness by assessing changes in students' attitudes and behaviors," were ranked (in order) significantly higher by students.
A key point to emphasize here is that while classifying students' performance levels based on current behavior might be useful in a practical context (for example to identify students in need of extra support), it is not our primary reason for using machine learning in the current study.
In fact, using ClassDojo is pretty straightforward: Teachers assign students an avatar and, based on their behavior throughout the day, dole out rewards or consequences (+1 for smiling, -1 for punching) from their laptop, tablet or smartphone.
Many schools have adopted effective programs such as School-Wide Positive Behavior Interventions and Supports (SWPBIS), a classroom management framework based on preventing behavior issues by building upon students' strengths, communicating clear expectations, and consistently modeling respectful behavior.
Mr. Benardo said students were selected for the trip based on their behavior and effort in class; the group includes four special-education students.
