Mathematical Sciences-Colloquium-Kang Xue (University of Florida) "Semi-Supervised Learning Method to Adjust Biased Item Difficulty Estimates Caused by Nonignorable Missingness in a Virtual Learning Environment"

Speaker: Kang Xue (University of Florida)

Title: Semi-Supervised Learning Method to Adjust Biased Item Difficulty Estimates Caused by Nonignorable Missingness in a Virtual Learning Environment

Abstract:
In data collected from virtual learning environments, nonignorable missing data patterns may impact the performance of applying psychometric models to item parameter and ability estimation. In this research, the factors related to missingness was explored and the biased estimates were adjusted using a semi-supervised learning method under 2 parameter item response theory (2PL-IRT) framework.