Zero-Frequency Cell Correction Strategies in Tetrachoric Correlation Estimation: Expanded Strategies and Multivariate Implications
Journal of Behavioral Data Science · 2026
DOI: 10.35566/jbds/choiwu ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Behavioral Data Science, 2026. Verfasst von Jeongwon Choi und 1 weiteren Autor:innen. Die Arbeit nennt 17 Literaturangaben.
Abstract (Originalsprache)
Zero-frequency cells pose a challenge for tetrachoric correlation estimation, but investigation of correction strategies remains limited. This study evaluates several zero-cell correction strategies, including different values to add, different ways to add the value, and the use of unadjusted versus adjusted thresholds in the second stage in the two-stage procedure. These strategies are examined across different correlation sizes and thresholds, to estimate a single tetrachoric correlation and extended to multivariate applications involving a tetrachoric correlation matrix and a confirmatory factor analysis model for binary data. Using multiple evaluation criteria, we show how these strategies perform differently across correlation sizes and the pattern of thresholds. This study also introduces ways to improve computational efficiency for tetrachoric correlation simulation studies that leverage the discrete structure to reduce redundant computations.
Zitieren
APA
Choi, J., & Wu, H. (2026). Zero-Frequency Cell Correction Strategies in Tetrachoric Correlation Estimation: Expanded Strategies and Multivariate Implications. Journal of Behavioral Data Science. https://doi.org/10.35566/jbds/choiwu
Vancouver
Choi J, Wu H. Zero-Frequency Cell Correction Strategies in Tetrachoric Correlation Estimation: Expanded Strategies and Multivariate Implications. Journal of Behavioral Data Science. 2026. doi:10.35566/jbds/choiwu
BibTeX
@article{Choi2026,
author = {Choi, Jeongwon and Wu, Hao},
title = {Zero-Frequency Cell Correction Strategies in Tetrachoric Correlation Estimation: Expanded Strategies and Multivariate Implications},
journal = {Journal of Behavioral Data Science},
year = {2026},
doi = {10.35566/jbds/choiwu}
}
EUVIMED