An IRTree Model for Aberrant Response and Missing Data in Standardized Tests

A 301.3 KB DOCX file by Fangbin Chen, last updated May 8, 2026, proposes a new Item Response Tree (IRTree) model. The model simultaneously accounts for rapid guessing, cheating, and nonresponse behaviors in standardized tests to improve parameter estimation. It is validated using two real datasets and simulation studies.

Use Cases

• Modeling examinee test-taking behaviors like rapid guessing and cheating based on the described IRTree framework.
• Improving the estimation of person and item parameters in standardized tests based on the proposed classification system.
• Simulating test data to validate psychometric models based on the described methodology.
• Comparing the performance of different aberrant response models based on the paper's evaluation.

Strengths

• Model provides classifications for examinee behaviors at both item and examinee levels, as described.
• It is the first model described to separate and simultaneously model guessing and cheating behaviors.
• License is CC-BY-4.0, permitting broad reuse with attribution.

Limitations

• The dataset is a 301.3 KB DOCX document, not a structured data file; actual data content is unknown.
• Row count and column-level documentation are absent; data quality and structure require manual inspection.
• The description focuses on the model methodology; the nature and scope of the two real datasets used are not detailed.

Provenance

Source

figshare

Collection Method

Likely contains research paper and possibly accompanying data or simulation results.

Time Range

null

Freshness

Last updated 2026-05-08 16:00:04

Geography

null