Research

Publications

Sufficient and Necessary Conditions for the Identifiability of DINA Models with Polytomous Responses, Mengqi Lin and Gongjun Xu. Psychometrika, 89(2), 717–740, 2024.

Preprints

Characterizing Identifiability in Boolean Factor Models, Mengqi Lin and Gongjun Xu. Major revision at Journal of the American Statistical Association.

Stochastic Sensitivity Analysis for Matched Observational Studies, Mengqi Lin, Colin Fogarty, and Gongjun Xu. Submitted.

Simultaneous Inference for False Discovery Proportions under Sensitivity Models for Observational Studies, Mengqi Lin and Colin Fogarty. Submitted.

Software

• idQ, Python package for checking Q-matrix identifiability in cognitive diagnosis models.

• SensitivityFDP, Python package for simultaneous inference on false discovery proportions in observational studies under sensitivity models.

• dwbh-adaptive_weights, R package for dependence-adjusted weighted Benjamini-Hochberg and general step-up procedures, with adaptive weighting.

Talks

• Characterizing Identifiability in Boolean Factor Models. JSM 2026, Boston; JSM 2025, Nashville.

• Controlling the False Discovery Proportion in Observational Studies with Hidden Bias. ICSA 2025, Taipei; ACIC 2025, Detroit; JSM 2024, Portland.

• Controlling the False Discovery Rate Under Dependency with the Adaptively Weighted BH Procedure. JSM 2023, Toronto; MCP 2022, Bremen.

• Identifiability of Cognitive Diagnosis Models with Polytomous Responses. IMPS 2023, Washington, D.C.

Awards

• Student Paper Award, Mental Health Statistics Section, American Statistical Association, 2026.

• Rackham Conference Travel Grant, University of Michigan, 2023, 2024, 2025.

• 1st prize poster award at MCP 2022, Bremen, Germany.

Service

• Journal referee: Statistica Sinica, Psychometrika, Journal of Educational and Behavioral Statistics, Scientific Reports.

• Volunteer, International Chinese Statistical Association (ICSA), 2023.