Route items flagged for misfit through content and psychometric review before deciding whether to deactivate them¶

Strategy · All strategies
Evidence · no claims cited

Description¶

This strategy operationalizes item fit analysis as a quality-control step in maintaining a Rasch-based assessment. After flagging items using point measure correlations below 0.2 and infit/outfit criteria, the report plots item characteristic curves for flagged items and states that "These items will receive both content and psychometric reviews before being deactivated." Flagging alone does not remove an item; review precedes any deactivation decision.

Design Implications¶

Context¶

Requirements¶

• Item-level response data sufficient to compute infit, outfit, and point measure correlations, with items having fewer than 300 responses excluded from the analysis.

Constraints¶

• The procedure was applied to MAP Growth K–2 Reading and Mathematics items whose difficulties were adjusted in the 2020 scale alignment study.

Target Learners¶

• K–2 students taking MAP Growth Reading and Mathematics assessments

Target Learning Goals¶

• Accurate measurement of early literacy and numeracy achievement on the Rasch-based MAP Growth K–2 scales

Related Strategies¶

• Use a two-step quality-indicator and decision-rule procedure that considers both high-quality and acceptable-quality research when establishing level of evidence

Examples¶

-

Key Sources¶

• He, W. (2020). MAP Growth K–2 item fit analysis: A follow-up study. NWEA. https://www.nwea.org/research/publication/map-growth-k-2-item-fit-analysis-study-a-follow-up-study/