Evaluating and Mitigating Second-Language Bias in LLM-Based Writing Feedback with Category-Resolved Counterfactual Invariance

Serena Lam; Samuel Carter; Sophia Lewis

doi:10.54097/fvhggk24

Authors

Serena Lam College of Education, Pennsylvania State University, USA
Samuel Carter Graduate School of Education, Rutgers University–New Brunswick, USA
Sophia Lewis Graduate School of Education, Rutgers University–New Brunswick, USA

DOI:

https://doi.org/10.54097/fvhggk24

Keywords:

Automated writing evaluation, Large language models, Algorithmic fairness, Counterfactual invariance, English language learners, Formative feedback

Abstract

Scorers built on large language models (LLMs) increasingly drive analytic feedback for English language learners (ELLs). Learner errors should lower the scores for conventions or grammar, but they can also spill over into dimensions that the rubric defines independently of them, such as cohesion or vocabulary, and redirect the feedback a learner receives. We audit an LLM-based analytic feedback model on 1,200 ELLIPSE essays with category-resolved counterfactuals, namely spelling, grammatical-morphology and preposition errors whose rates are re-estimated from JFLEG and TOEFL-Spell, and with authentic misspellings corrected in the students’ own essays. Added misspellings lower the predicted cohesion, syntax and vocabulary scores by about half as much as conventions and change the rounded scores of about a quarter of the learners. We propose CR-CAIR, which extends construct-aligned invariance regularization with a rubric-derived category–trait construct map and a sensitivity-anchoring term while keeping a closed-form solution. Over five splits, CR-CAIR reduces spill-over on construct-irrelevant cells by 86% and the corresponding flip rate from 12.2% to 4.2%, keeps the legitimate sensitivity of conventions, grammar and phraseology, nearly halves misdirected feedback priorities and transfers to authentic misspellings (−79% spill-over), with no significant change in agreement (QWK 0.649 vs. 0.654). Task-agnostic debiasing removes legitimate sensitivity, and trait-level invariance increases misdirected feedback.

Downloads

Download data is not yet available.

References

[1] Taghipour, K., & Ng, H. T. (2016). A neural approach to automated essay scoring. In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing (EMNLP) (pp. 1882-1891). Association for Computational Linguistics. https://doi.org/10.18653/v1/D16-1193

[2] Li, S., & Ng, V. (2024). Automated essay scoring: A reflection on the state of the art. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP) (pp. 17876-17888). Association for Computational Linguistics. https://doi.org/10.18653/v1/2024.emnlp-main.991

[3] Mizumoto, A., & Eguchi, M. (2023). Exploring the potential of using an AI language model for automated essay scoring. Research Methods in Applied Linguistics, 2(2), Article 100050. https://doi.org/10.1016/j.rmal.2023.100050

[4] Yancey, K. P., Laflair, G., Verardi, A., & Burstein, J. (2023). Rating short L2 essays on the CEFR scale with GPT-4. In Proceedings of the 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA) (pp. 576-584). Association for Computational Linguistics. https://doi.org/10.18653/v1/2023.bea-1.49

[5] Zhou, F. (2026). Moment-constrained side-force allocation for wind-assisted merchant ships: Hydrodynamic performance of retractable anti-leeway fins and flap rudders. Innovation and Technology Studies, 3(1), 27-41.

[6] Steiss, J., Tate, T., Graham, S., Cruz, J., Hebert, M., Wang, J., Moon, Y., Tseng, W., Warschauer, M., & Olson, C. B. (2024). Comparing the quality of human and ChatGPT feedback of students' writing. Learning and Instruction, 91, Article 101894. https://doi.org/10.1016/j.learninstruc.2024.101894

[7] Crossley, S. A., Tian, Y., Baffour, P., Franklin, A., Kim, M., Morris, W., Choi, J., Picou, A., & Stover, T. (2023). Measuring second language proficiency using the English Language Learner Insight, Proficiency and Skills Evaluation (ELLIPSE) corpus. International Journal of Learner Corpus Research, 9(2), 248-269. https://doi.org/10.1075/ijlcr.22026.cro

[8] Liang, W., Yuksekgonul, M., Mao, Y., Wu, E., & Zou, J. (2023). GPT detectors are biased against non-native English writers. Patterns, 4(7), Article 100779. https://doi.org/10.1016/j.patter.2023.100779

[9] Chen, J., Han, X., Zhang, X., & Yang, Y. (2023). Exact concept-level explanations for early detection of financial misstatements. Eurasia Journal of Science and Technology, 5(2), 34-48.

[10] Williamson, D. M., Xi, X., & Breyer, F. J. (2012). A framework for evaluation and use of automated scoring. Educational Measurement: Issues and Practice, 31(1), 2-13. https://doi.org/10.1111/j.1745-3992.2011.00223.x

[11] Bridgeman, B., Trapani, C., & Attali, Y. (2012). Comparison of human and machine scoring of essays: Differences by gender, ethnicity, and country. Applied Measurement in Education, 25(1), 27-40. https://doi.org/10.1080/08957347.2012.635502

[12] Loukina, A., Madnani, N., & Zechner, K. (2019). The many dimensions of algorithmic fairness in educational applications. In Proceedings of the 14th Workshop on Innovative Use of NLP for Building Educational Applications (BEA) (pp. 1-10). Association for Computational Linguistics. https://doi.org/10.18653/v1/W19-4401

[13] Kusner, M. J., Loftus, J., Russell, C., & Silva, R. (2017). Counterfactual fairness. Advances in Neural Information Processing Systems, 30, 4066-4076.

[14] Garg, S., Perot, V., Limtiaco, N., Taly, A., Chi, E. H., & Beutel, A. (2019). Counterfactual fairness in text classification through robustness. In Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society (AIES) (pp. 219-226). ACM. https://doi.org/10.1145/3306618.3317950

[15] Fang, Q. (2024). Bias-aware large language models for equitable educational feedback and assessment. Trends in Social Sciences and Humanities Research, 2(1), 41-50. https://doi.org/10.61784/tsshr4253

[16] Baker, R. S., & Hawn, A. (2022). Algorithmic bias in education. International Journal of Artificial Intelligence in Education, 32, 1052-1092. https://doi.org/10.1007/s40593-021-00285-9

[17] Blodgett, S. L., Barocas, S., Daumé III, H., & Wallach, H. (2020). Language (technology) is power: A critical survey of "bias" in NLP. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL) (pp. 5454-5476). Association for Computational Linguistics. https://doi.org/10.18653/v1/2020.acl-main.485

[18] Wang, J., Ge, Y., Liu, J., & Jiang, B. (2024). Early warning of liquidity stress in U.S. banks with seasonally aligned frailty hazards and delayed-feedback conformal alarms. Journal of Trends in Finance and Economics, 1(1), 85-100.

[19] Elazar, Y., & Goldberg, Y. (2018). Adversarial removal of demographic attributes from text data. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing (EMNLP) (pp. 11-21). Association for Computational Linguistics. https://doi.org/10.18653/v1/D18-1002

[20] Ravfogel, S., Elazar, Y., Gonen, H., Twiton, M., & Goldberg, Y. (2020). Null it out: Guarding protected attributes by iterative nullspace projection. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL) (pp. 7237-7256). Association for Computational Linguistics. https://doi.org/10.18653/v1/2020.acl-main.647

[21] Liang, P. P., Li, I. M., Zheng, E., Lim, Y. C., Salakhutdinov, R., & Morency, L.-P. (2020). Towards debiasing sentence representations. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL) (pp. 5502-5515). Association for Computational Linguistics. https://doi.org/10.18653/v1/2020.acl-main.488

[22] Napoles, C., Sakaguchi, K., & Tetreault, J. (2017). JFLEG: A fluency corpus and benchmark for grammatical error correction. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics (EACL), Vol. 2: Short Papers (pp. 229-234). Association for Computational Linguistics.

[23] Flor, M., Fried, M., & Rozovskaya, A. (2019). A benchmark corpus of English misspellings and a minimally-supervised model for spelling correction. In Proceedings of the 14th Workshop on Innovative Use of NLP for Building Educational Applications (BEA) (pp. 76-86). Association for Computational Linguistics. https://doi.org/10.18653/v1/W19-4407

[24] Felice, M., & Yuan, Z. (2014). Generating artificial errors for grammatical error correction. In Proceedings of the Student Research Workshop at the 14th Conference of the European Chapter of the Association for Computational Linguistics (pp. 116-126). Association for Computational Linguistics. https://doi.org/10.3115/v1/E14-3013

[25] Honnibal, M., Montani, I., Van Landeghem, S., & Boyd, A. (2020). spaCy: Industrial-strength natural language processing in Python. Zenodo. https://doi.org/10.5281/zenodo.1212303

[26] Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., Levy, O., Lewis, M., Zettlemoyer, L., & Stoyanov, V. (2019). RoBERTa: A robustly optimized BERT pretraining approach. arXiv, arXiv:1907.11692. https://doi.org/10.48550/arXiv.1907.11692

[27] Chen, J., Zhang, X., & Han, X. (2023). Explainable identification of tax-related financial reporting anomalies in U.S. public companies: Cross-fitted tax-consistency residuals and Owen-value Shapley propagation. World Journal of Economics and Business Research, 1(1), 33-45.

[28] Cohen, J. (1968). Weighted kappa: Nominal scale agreement with provision for scaled disagreement or partial credit. Psychological Bulletin, 70(4), 213-220. https://doi.org/10.1037/h0026256

[29] Chen, J., Liu, J., Liang, Y., & Zhou, M. (2026). KE-MLLM: A knowledge-enhanced multi-sensor learning framework for explainable fake review detection. Applied Sciences, 16, Article 2909. https://doi.org/10.3390/app16062909

[30] Zeng, Z., Zhang, Y., Li, H., Wang, J., & Chen, X. (2026). Adaptive robust watermarking for large language models via dynamic token embedding perturbation. IEEE Access, 14, 1-15. https://doi.org/10.1109/ACCESS.2026.3653833