The initial version of autoFC accepted by
Applied Psychological Measurement (Li et al., 2022).
Designed for reducing the cognitive load of manually constructing forced-choice blocks based on multiple criteria (e.g., social desirability matching, item loading, trait pairing) Includes the core functionality for automatically pair statements in each forced-choice block.
The version used for the paper accepted by Organizational Research Methods (Li et al., 2025).
Improved functionality allowing users to construct pre-defined forced-choice blueprints and full FC based on these blueprints
Wrapper functions for more user-friendly response conversion,
data pre-processing and scoring using the thurstonianIRT
package.
Various diagnostic functions added.
irrCAC package to resolve
the package archiving issue.This major update introduces critical performance, usability, and mathematical enhancements. These features are added to allow for MUCH easier and convenient use of FC scales, than ever before.
Thurstonian Factor Model (TFM) and Thurstonian IRT (IRT) Support:
Native syntax generation and scoring for highly stable first-order and
second-order models in lavaan and Mplus, completely bypassing tedious
manual constraint declarations. No dependencies on
thurstonianIRT package is needed now.
Vectorized, C-Level Algorithmic Speedups: Restructured simulated annealing energy calculators and agreement metrics (bp.coeff.raw, gwet.ac1.raw) to achieve significant performance improvements.
High-Performance Stan Integration: A unified Stan pipeline featuring within-chain parallel computing (reduce_sum) and automatic logical dependency reduction (Heister, Doebler, & Frick, 2025) to easily scale to larger blocks.
Blazing Fast Scoring: Fast, vectorized MAP scoring in R using gradients to score participants and calculate Standard Errors in seconds (R and Mplus models).
Enhanced Blueprint Functions and Support for Global Optimization: Users have always been wondering whether we can optimize FC measures (on block level) while strictly following a certain designation of trait pair and/or keying distribution or strictly aligning with a pre-defined blueprint (on global scale level). We have now implemented easy-to-use functionality serving exactly for this purpose.
generate_tirt_mplus_syntax() function
where originally the designation of trait_col and
key_col parameters have no effect.Li, M., Sun, T., & Zhang, B. (2022). AutoFC: An R package for automatic item pairing in forced-choice test construction. Applied Psychological Measurement, 46(1), 70-72. https://doi.org/10.1177/01466216211051726
Li, M., Zhang, B., Li, L., Sun, T., & Brown, A. (2025). Mixed-keying or desirability-matching in the construction of forced-choice measures? An empirical investigation and practical recommendations. Organizational Research Methods, 28(2), 296-329. https://doi.org/10.1177/10944281241229784