Grow the MCD item bank during operation using online calibration embedded alongside scored items¶

Strategy · All strategies
Evidence · no claims cited

Description¶

To expand from 14 to 35 planned learning objectives without pausing testing, the authors recommend online calibration: "the system embeds new, unscored items alongside the operational item bank, collects real student responses to them during normal use, and estimates their DINA parameters from those responses." Once parameters stabilize, the item joins the scored bank, letting the bank grow while the MCD stays in use. They also recommend adding existing RBAs for uncovered topics and using accumulating EMCS responses to compare DINA against G-DINA and hierarchical alternatives.

Design Implications¶

Context¶

Requirements¶

• A live delivery platform with sufficient response volume for parameter estimates to stabilize
• Embedding unscored items during normal assessment use rather than a separate pretesting phase

Constraints¶

Target Learners¶

• College students in introductory mechanics courses

Target Learning Goals¶

• Expanding diagnostic coverage to rotational mechanics, mathematical reasoning, and additional energy-specific objectives

Related Strategies¶

• Use CDP-simulated examinees for early item calibration, data augmentation, diagnostic item design, and fairness review before human field testing

Examples¶

-

Key Sources¶

• Vy Le, Jayson M. Nissen, Jason W. Morphew, Hua Hua Chang, and Ben Van Dusen. (2026). Mechanics Cognitive Diagnostic: Testing Fine-Grained Learning Objectives in Introductory Physics. https://arxiv.org/abs/2609.09584