Cengiz Zopluoglu: Enhanced Deterministic Gated Item Response Model to Simultaneously Identify Compromised Items and Examinees with Item Preknowledge Using Response Accuracy Data

Cengiz Zopluoglu

Enhanced Deterministic Gated Item Response Model to Simultaneously Identify Compromised Items and Examinees with Item Preknowledge Using Response Accuracy Data

item response theory Stan R item preknowledge detecting test misconduct 2024

This post introduces the Enhanced Deterministic Gated Item Response Model (DG-IRT), a new model developed for detecting both compromised test items and examinees who might have prior knowledge—without needing to know which items are compromised ahead of time. I walk through the model setup, simulation, and a complete Stan implementation, showing how to estimate everything from scratch using only response accuracy data.

Citation

For attribution, please cite this work as

Zopluoglu (2024, Dec. 6). Cengiz Zopluoglu: Enhanced Deterministic Gated Item Response Model to Simultaneously Identify Compromised Items and Examinees with Item Preknowledge Using Response Accuracy Data. Retrieved from https://github.com/czopluoglu/website/tree/master/docs/posts/duolingo_dgirt/

BibTeX citation

@misc{zopluoglu2024enhanced,
  author = {Zopluoglu, Cengiz},
  title = {Cengiz Zopluoglu: Enhanced Deterministic Gated Item Response Model to Simultaneously Identify Compromised Items and Examinees with Item Preknowledge Using Response Accuracy Data},
  url = {https://github.com/czopluoglu/website/tree/master/docs/posts/duolingo_dgirt/},
  year = {2024}
}