The Impact of Explicit Strategy Instruction on Problem-solving Behaviors across Intelligent Tutoring Systems

Explicit instruction of a problem-solving strategy improved students’ performance in the domain where it was taught and in a second domain where it was not taught. Since the two domains, probability and physics, share no overlapping domain principles, it seems likely that the problem-solving strategy itself was transferred from one domain to the other. We analyzed the computer logs to identify the strategies employed in the physics domain. On hard problems the Strategy group did in fact employ the taught strategy in question. However on easier problems they appeared to abandon the strategy in favor of more sophisticated strategies. We conjecture that explicit strategy instruction increased students’ focus on domain principle acquisition, which in turn accelerated their learning to the point where they no longer needed to use the taught strategy and instead could use more expert-like strategies. However, when they got stuck, they fell back on the taught strategy. That is, explicit strategy instruction appears to act as surprisingly effective scaffolding for domain principle acquisition.