Testing the Robustness of Inquiry Practices Once Scaffolding Is Removed

摘要

Intelligent tutoring systems (ITS) with simulated and virtual labs have been designed to enhance students' science knowledge, including content and inquiry practices; some systems do this via real-time scaffolding. Prior studies have demonstrated that scaffolding can benefit students' learning and performance. The present study aims to examine the robustness of scaffolding, delivered by a pedagogical agent by providing scaffolding on one activity, removing it, and then evaluating students' inquiry performance both over multiple time periods (in 40 days, 80 days, and 170 days) and across different topics, thereby addressing far transfer. 107 middle school students in grade 6 received adaptive scaffolding on the first inquiry topic (i.e. Animal Cell) in the intelligent tutoring system, Inq-ITS. Then they received no scaffolding for three topics, namely, Plant Cell, Genetics, and Natural Selection. Results showed that after removing scaffolding, students demonstrated continued growth of inquiry performance from time 1 to time 2, to time 3, and to time 4 for the practices of hypothesizing and collecting data, as well as from time 1 to time 2 and to time 4 for the practice of warranting claims. This pattern was not found in students' performance on the practice of interpreting data. These findings have implications for designers and researchers regarding the design of scaffolds for the NGSS' inquiry practices so that they can be effectively transferred. These data also point to the need for additional work to address content practice interactions.