Beyond Access: Guided LLM Scaffolding for Independent Learning in Undergraduate Statistics
The episode discusses a paper titled "Beyond Access: Guided LLM Scaffolding for Independent Learning in Undergraduate Statistics." The hosts analyze an experiment comparing three student groups: no AI access, unrestricted AI access, and guided AI access. They conclude that structured guidance—teaching students to ask for hints and explanations rather than answers—led to better independent learning performance when the AI was removed.