2025 ASEE Annual Conference & Exposition

Traditional classroom observation often rely on human observers in a process that can be labor intensive, time-consuming, and may not fully capture individual student experiences. Sensing technologies, such as Bluetooth Low Energy (BLE) beacons, offer a promising alternative by collecting continuous, real-time data on student proximity and interactions. This paper evaluates the use of BLE beacons as part of a real-time social interaction capture system, called IDEAS. In a laboratory setting, the relationship between beacon signal strength (RSSI), distance, and orientation was examined to validate a proximity detection metric. A preschool classroom study further tested the ecological validity of the real-time location system by comparing interactions detected by the automated system with those recorded by a traditional researcher-led method. In order to align the differing sampling methods of IDEAS and the traditional researcher-led method, we developed an algorithm to down-sample the beacon data. The results suggest a partial alignment between beacon-detected interactions and the ones detected via traditional observations. The limited correspondence could be due to signal loss or inconsistencies in human observations. Incorporating additional sensor modalities, such as audio from wearable recorders, may enhance the system's accuracy. These findings highlight the potential for systems like IDEAS to offer scalable methods for capturing peer interactions and engagement through multi-modal sensor data. Building on approaches used in early childhood education research, incorporating sensing technologies into engineering education could provide richer, continuous insights into student collaboration and classroom dynamics.