The Help Tutor improved students' in-tutor help-seeking behavior, but only for a subset of help-seeking action types¶

Claim · All claims
Evidence · 1 study · causal r2 · q3
Currency · newest study 2008 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i? Students under the Help Tutor sought help more appropriately as measured by conformity to the help-seeking model, but the improvement held only for a subset of action types and not for first actions on new steps. → Koedinger 2008

Evidence¶

Koedinger 2008¶

Koedinger, K., Aleven, V., Roll, I., & Baker, R. (2008). In vivo experiments on whether supporting metacognition in intelligent tutoring systems yields robust learning. https://learnlab.org/

q3 · i? · causal · r2

First of two classroom experiments comparing the standard Geometry Cognitive Tutor to a version with the integrated Help Tutor, measuring conformity of help-seeking actions to the normative model. Improvement was limited to "only a subset of help-seeking action types"; no effect size printed.

"there was evidence that students sought help more appropriately under the Help Tutor's tutelage, as measured by the percentage of actions that conformed to the help-seeking model. However, the improvement was seen with respect to only a subset of help-seeking action types."

Discussion¶

Related Claims¶

• Students' help-seeking error frequencies correlate strongly across different Cognitive Tutor units and cohorts (r = 0.89)
• Single instances of help-seeking actions (answer requested, explanation requested) carry more predictive information than single correct or open responses (ISA) — related
• A warning message about potential AI mistakes increases seventh-graders' hint requests in a math intelligent tutoring system — related