{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T19:06:10Z","timestamp":1785524770044,"version":"3.56.0"},"reference-count":75,"publisher":"Public Library of Science (PLoS)","issue":"7","license":[{"start":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T00:00:00Z","timestamp":1785456000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>Recent developments in the field of artificial intelligence and machine learning allow the wide application of large language models for the evaluation of written text and other non-numerical data. When applied in the context of psychological and educational assessments, such models can be used for assigning scores to essays and other types of responses. In contrast to classical tests, essays do not consist of test items, which leads to specific challenges in the evaluation of testing standards for scores obtained from AI models that differ from those observed for classical ability tests and personality questionnaires. To address these challenges, we discuss the evaluation of validity, fairness, and reliability for scores obtained from models of artificial intelligence in the context of automated essay scoring. We review existing methods, propose new methods, and further illustrate the reviewed methods with an empirical example based on the Hewlett Foundation data set on automated essay scoring. By applying the proposed framework to an evaluation based on a DistilBERT model, we find the model to be robust with sufficiently high internal consistency (Spearman-Brown coefficients in the range from\u2009.77 to\u2009.92). We further found empirical evidence for the validity of the evaluation model, but also indications for violations of fairness when comparing the human and AI scores across different topics. This study provides a standardized, replicable toolkit for researchers and practitioners to evaluate the psychometric quality of AI-based assessments.<\/jats:p>","DOI":"10.1371\/journal.pone.0354680","type":"journal-article","created":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T17:41:16Z","timestamp":1785519676000},"page":"e0354680","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Testing standards for AI-based scores in automated essay scoring"],"prefix":"10.1371","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-8900-2106","authenticated-orcid":true,"given":"Rudolf","family":"Debelak","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4994-9519","authenticated-orcid":true,"given":"Matthias","family":"Ziegler","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,7,31]]},"reference":[{"key":"pone.0354680.ref001","unstructured":"OpenAI. GPT-4 Technical Report. 2023."},{"key":"pone.0354680.ref002","unstructured":"OpenAI. GPT-4o. https:\/\/openai.com\/index\/hello-gpt-4o\/ 2024."},{"key":"pone.0354680.ref003","unstructured":"Google AI. Introducing Gemini: Our Next Generation Model. https:\/\/ai.googleblog.com\/ 2023."},{"key":"pone.0354680.ref004","doi-asserted-by":"crossref","first-page":"903077","DOI":"10.3389\/frai.2022.903077","article-title":"The interactive reading task: Transformer-based automatic item generation","volume":"5","author":"Y Attali","year":"2022","journal-title":"Front Artif Intell"},{"key":"pone.0354680.ref005","first-page":"100161","article-title":"Automated reading passage generation with OpenAI\u2019s large language model","volume":"5","author":"U Bezirhan","year":"2023","journal-title":"Computers and Education: Artificial Intelligence"},{"issue":"2","key":"pone.0354680.ref006","doi-asserted-by":"crossref","first-page":"16","DOI":"10.1111\/emip.12602","article-title":"Transforming Assessment: The Impacts and Implications of Large Language Models and Generative AI","volume":"43","author":"J Hao","year":"2024","journal-title":"Educational Measurement"},{"issue":"1","key":"pone.0354680.ref007","article-title":"Combining machine translation and automated scoring in international large-scale assessments","volume":"12","author":"JY Jung","year":"2024","journal-title":"Large-scale Assess Educ"},{"key":"pone.0354680.ref008","doi-asserted-by":"crossref","DOI":"10.1201\/9781003054313","volume-title":"Computational aspects of psychometric methods: With R","author":"P Martinkov\u00e1","year":"2023"},{"issue":"19","key":"pone.0354680.ref009","doi-asserted-by":"crossref","DOI":"10.3390\/math11194104","article-title":"Item difficulty prediction using item text features: comparison of predictive performance across machine-learning algorithms","volume":"11","author":"L \u0160t\u011bp\u00e1nek","year":"2023","journal-title":"Mathematics"},{"issue":"4","key":"pone.0354680.ref010","doi-asserted-by":"crossref","first-page":"847","DOI":"10.1007\/s11336-018-9608-y","article-title":"Automated Item Generation with Recurrent Neural Networks","volume":"83","author":"M von Davier","year":"2018","journal-title":"Psychometrika"},{"issue":"2","key":"pone.0354680.ref011","first-page":"55","article-title":"Automated distractor generation for fill-in-the-blank items using a prompt-based learning approach","volume":"65","author":"J Zu","year":"2023","journal-title":"Psychological Test and Assessment Modeling"},{"issue":"1","key":"pone.0354680.ref012","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1111\/jedm.12129","article-title":"Computational Psychometrics in Support of Collaborative Educational Assessments","volume":"54","author":"AA von Davier","year":"2017","journal-title":"J Educational Measurement"},{"key":"pone.0354680.ref013","volume-title":"Computational psychometrics: New methodologies for a new generation of digital learning and assessment: With examples in R and Python","author":"AA von Davier","year":"2022"},{"key":"pone.0354680.ref014","doi-asserted-by":"crossref","unstructured":"Bulut O, Beiting-Parrish M, Casabianca JM, Slater SC, Jiao H, Song D. The rise of artificial intelligence in educational measurement: opportunities and ethical challenges. 2024. https:\/\/arxiv.org\/abs\/2406.18900","DOI":"10.59863\/MIQL7785"},{"key":"pone.0354680.ref015","doi-asserted-by":"crossref","unstructured":"Burstein J, LaFlair GT. Where assessment validation and responsible AI meet. https:\/\/arxiv.org\/abs\/2411.02577 2024.","DOI":"10.32038\/ltrq.2025.50.09"},{"key":"pone.0354680.ref016","doi-asserted-by":"crossref","first-page":"1474019","DOI":"10.3389\/frai.2024.1474019","article-title":"A generative AI-driven interactive listening assessment task","volume":"7","author":"A Runge","year":"2024","journal-title":"Front Artif Intell"},{"issue":"3","key":"pone.0354680.ref017","doi-asserted-by":"crossref","first-page":"288","DOI":"10.1111\/jedm.12333","article-title":"Validity Arguments for AI\u2010Based Automated Scores: Essay Scoring as an Illustration","volume":"59","author":"S Ferrara","year":"2022","journal-title":"J Educational Measurement"},{"key":"pone.0354680.ref018","doi-asserted-by":"crossref","unstructured":"Harris DJ. AI: Can You Help Address This Issue?. Educational Measurement: Issues and Practice. 2024. doi: 10.1111\/emip.12655","DOI":"10.1111\/emip.12655"},{"key":"pone.0354680.ref019","doi-asserted-by":"crossref","unstructured":"Iliescu D, Greiff S, Ziegler M, Fokkema M. Artificial Intelligence, Machine Learning, and Other Demons. 2022.","DOI":"10.1027\/1015-5759\/a000713"},{"issue":"3","key":"pone.0354680.ref020","doi-asserted-by":"crossref","first-page":"338","DOI":"10.1111\/jedm.12335","article-title":"Psychometric Methods to Evaluate Measurement and Algorithmic Bias in Automated Scoring","volume":"59","author":"MS Johnson","year":"2022","journal-title":"J Educational Measurement"},{"issue":"13","key":"pone.0354680.ref021","doi-asserted-by":"crossref","DOI":"10.1073\/pnas.2215907120","article-title":"The debate over understanding in AI\u2019s large language models","volume":"120","author":"M Mitchell","year":"2023","journal-title":"Proc Natl Acad Sci U S A"},{"key":"pone.0354680.ref022","volume-title":"Standards for educational and psychological testing","author":"American Educational Research Association, American Psychological Association, National Council on Measurement in Education","year":"2014"},{"key":"pone.0354680.ref023","volume-title":"Guidelines for technology-based assessment","author":"International Test Commission, Association of Test Publishers","year":"2022"},{"key":"pone.0354680.ref024","volume-title":"Psychometrics","author":"CR Rao","year":"2007"},{"key":"pone.0354680.ref025","volume-title":"Statistical theories of mental test scores","author":"FM Lord","year":"1968"},{"key":"pone.0354680.ref026","doi-asserted-by":"crossref","DOI":"10.1201\/9781315374512","volume-title":"Handbook of Item Response Theory \u2013 Three Volume Set","author":"W van der Linden","year":"2016"},{"key":"pone.0354680.ref027","unstructured":"Devlin J, Chang MW, Lee K, Toutanova K. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. In: 2019. https:\/\/www.aclweb.org\/anthology\/N19-1423"},{"issue":"3","key":"pone.0354680.ref028","article-title":"Automated essay scoring with e-rater\u00ae v.2","volume":"4","author":"Y Attali","year":"2006","journal-title":"The Journal of Technology, Learning and Assessment"},{"issue":"1","key":"pone.0354680.ref029","doi-asserted-by":"crossref","first-page":"2","DOI":"10.1111\/j.1745-3992.2011.00223.x","article-title":"A Framework for Evaluation and Use of Automated Scoring","volume":"31","author":"DM Williamson","year":"2012","journal-title":"Educational Measurement"},{"key":"pone.0354680.ref030","unstructured":"Sanh V, Debut L, Chaumond J, Wolf T. DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter. 2020."},{"issue":"8","key":"pone.0354680.ref031","doi-asserted-by":"crossref","first-page":"1277","DOI":"10.1037\/apl0001082","article-title":"How well can an AI chatbot infer personality? Examining psychometric properties of machine-inferred personality scores","volume":"108","author":"J Fan","year":"2023","journal-title":"J Appl Psychol"},{"issue":"5","key":"pone.0354680.ref032","doi-asserted-by":"crossref","first-page":"967","DOI":"10.1037\/met0000516","article-title":"Updated guidelines on selecting an intraclass correlation coefficient for interrater reliability, with applications to incomplete observational designs","volume":"29","author":"D Ten Hove","year":"2024","journal-title":"Psychol Methods"},{"issue":"7","key":"pone.0354680.ref033","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1371\/journal.pone.0219854","article-title":"Intraclass correlation \u2013 A discussion and demonstration of basic features","volume":"14","author":"D Liljequist","year":"2019","journal-title":"PLOS ONE"},{"issue":"2","key":"pone.0354680.ref034","doi-asserted-by":"crossref","first-page":"155","DOI":"10.1016\/j.jcm.2016.02.012","article-title":"A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research","volume":"15","author":"TK Koo","year":"2016","journal-title":"J Chiropr Med"},{"issue":"3","key":"pone.0354680.ref035","doi-asserted-by":"crossref","first-page":"349","DOI":"10.3102\/10769986221150517","article-title":"Assessing Inter-rater Reliability With Heterogeneous Variance Components Models: Flexible Approach Accounting for Contextual Variables","volume":"48","author":"P Martinkov\u00e1","year":"2023","journal-title":"Journal of Educational and Behavioral Statistics"},{"key":"pone.0354680.ref036","first-page":"181","article-title":"Validity and reliability of automated essay scoring","volume-title":"Validity and reliability of automated essay scoring","author":"Y Attali","year":"2013"},{"issue":"5","key":"pone.0354680.ref037","first-page":"238","article-title":"The imminence of\u2026 grading essays by computer","volume":"47","author":"EB Page","year":"1966","journal-title":"Phi Delta Kappan"},{"key":"pone.0354680.ref038","doi-asserted-by":"crossref","first-page":"72","DOI":"10.1016\/j.asw.2017.10.002","article-title":"College student perceptions of writing errors, text quality, and author characteristics","volume":"34","author":"AC Johnson","year":"2017","journal-title":"Assessing Writing"},{"issue":"21","key":"pone.0354680.ref039","doi-asserted-by":"crossref","DOI":"10.3390\/math12213416","article-title":"Hybrid approach to automated essay scoring: integrating deep learning embeddings with handcrafted linguistic features for improved accuracy","volume":"12","author":"M Faseeh","year":"2024","journal-title":"Mathematics"},{"key":"pone.0354680.ref040","volume-title":"Deep Learning","author":"I Goodfellow","year":"2016"},{"issue":"3","key":"pone.0354680.ref041","doi-asserted-by":"crossref","first-page":"36","DOI":"10.1111\/emip.12036","article-title":"Managing What We Can Measure: Quantifying the Susceptibility of Automated Scoring Systems to Gaming Behavior","volume":"33","author":"D Higgins","year":"2014","journal-title":"Educational Measurement"},{"issue":"2","key":"pone.0354680.ref042","doi-asserted-by":"crossref","first-page":"103","DOI":"10.1016\/S0747-5632(01)00052-8","article-title":"Stumping e-rater:challenging the validity of automated essay scoring","volume":"18","author":"DE Powers","year":"2002","journal-title":"Computers in Human Behavior"},{"issue":"3","key":"pone.0354680.ref043","doi-asserted-by":"crossref","first-page":"412","DOI":"10.1177\/10944281251346404","article-title":"Reliability Evidence for AI-Based Scores in Organizational Contexts: Applying Lessons Learned From Psychometrics","volume":"29","author":"AB Speer","year":"2025","journal-title":"Organizational Research Methods"},{"issue":"2","key":"pone.0354680.ref044","doi-asserted-by":"crossref","first-page":"81","DOI":"10.1037\/h0046016","article-title":"Convergent and discriminant validation by the multitrait-multimethod matrix","volume":"56","author":"DT CampbelL","year":"1959","journal-title":"Psychol Bull"},{"issue":"2","key":"pone.0354680.ref045","doi-asserted-by":"crossref","first-page":"219","DOI":"10.1007\/s11205-011-9843-4","article-title":"Validity and the Consequences of Test Interpretation and Use","volume":"103","author":"AM Hubley","year":"2011","journal-title":"Soc Indic Res"},{"issue":"2","key":"pone.0354680.ref046","doi-asserted-by":"crossref","first-page":"190","DOI":"10.1177\/1088868318772990","article-title":"Using Machine Learning to Advance Personality Assessment and Theory","volume":"23","author":"W Bleidorn","year":"2019","journal-title":"Pers Soc Psychol Rev"},{"issue":"1","key":"pone.0354680.ref047","doi-asserted-by":"crossref","first-page":"127","DOI":"10.7334\/psicothema2013.258","article-title":"Validity evidence based on testing consequences","volume":"26","author":"S Lane","year":"2014","journal-title":"Psicothema"},{"key":"pone.0354680.ref048","unstructured":"Molnar C. Interpretable Machine Learning. 2nd ed. 2022."},{"key":"pone.0354680.ref049","unstructured":"Hvitfeldt E, Pedersen TL, Benesty M. lime: Local Interpretable Model-Agnostic Explanations. 2022. https:\/\/CRAN.R-project.org\/package=lime"},{"key":"pone.0354680.ref050","first-page":"307","article-title":"A value for n-person games","volume-title":"A Value for n-Person Games","author":"LS Shapley","year":"1953"},{"key":"pone.0354680.ref051","unstructured":"Abadi M, Agarwal A, Barham P, Brevdo E, Chen Z, Citro C. TensorFlow: Large-Scale Machine Learning on Heterogeneous Systems. 2015. https:\/\/www.tensorflow.org\/"},{"key":"pone.0354680.ref052","doi-asserted-by":"crossref","unstructured":"Ding Y, Riordan B, Horbach A, Cahill A, Zesch T. Don\u2019t take \u201cnswvtnvakgxpm\u201d for an answer \u2013The surprising vulnerability of automatic content scoring systems to adversarial input. In: Proceedings of the 28th International Conference on Computational Linguistics, 2020. 882\u201392. https:\/\/doi.org\/10.18653\/v1\/2020.coling-main.76","DOI":"10.18653\/v1\/2020.coling-main.76"},{"issue":"1","key":"pone.0354680.ref053","doi-asserted-by":"crossref","first-page":"38","DOI":"10.1037\/1082-989X.8.1.38","article-title":"Separating trait effects from trait-specific method effects in multitrait-multimethod models: a multiple-indicator CT-C(M-1) model","volume":"8","author":"M Eid","year":"2003","journal-title":"Psychol Methods"},{"issue":"3","key":"pone.0354680.ref054","doi-asserted-by":"crossref","first-page":"88","DOI":"10.1027\/1614-2241.5.3.88","article-title":"A CTC(M\u22121) Model for Different Types of Raters","volume":"5","author":"FW Nussbeck","year":"2009","journal-title":"Methodology"},{"issue":"1","key":"pone.0354680.ref055","article-title":"A Conceptual Framework for Investigating and Mitigating Machine-Learning Measurement Bias (MLMB) in Psychological Assessment","volume":"5","author":"L Tay","year":"2022","journal-title":"Advances in Methods and Practices in Psychological Science"},{"issue":"3","key":"pone.0354680.ref056","doi-asserted-by":"crossref","first-page":"248","DOI":"10.1037\/1082-989X.2.3.248","article-title":"Invariance in measurement and prediction: Their relationship in the single-factor case","volume":"2","author":"RE Millsap","year":"1997","journal-title":"Psychological Methods"},{"issue":"4","key":"pone.0354680.ref057","doi-asserted-by":"crossref","first-page":"461","DOI":"10.1007\/s11336-007-9039-7","article-title":"Invariance in Measurement and Prediction Revisited","volume":"72","author":"RE Millsap","year":"2007","journal-title":"Psychometrika"},{"key":"pone.0354680.ref058","volume-title":"Fairness and Machine Learning: Limitations and Opportunities","author":"S Barocas","year":"2023"},{"issue":"1","key":"pone.0354680.ref059","doi-asserted-by":"crossref","first-page":"37","DOI":"10.1027\/1866-5888\/a000287","article-title":"Pitfalls of Machine Learning-Based Personnel Selection","volume":"21","author":"D Goretzko","year":"2022","journal-title":"Journal of Personnel Psychology"},{"key":"pone.0354680.ref060","unstructured":"Bolukbasi T, Chang KW, Zou J, Saligrama V, Kalai A. Man is to computer programmer as woman is to homemaker? Debiasing word embeddings. In: 2016. https:\/\/arxiv.org\/abs\/1607.06520"},{"issue":"1","key":"pone.0354680.ref061","doi-asserted-by":"crossref","first-page":"30064","DOI":"10.1038\/s41598-024-79208-2","article-title":"Examining the responsible use of zero-shot AI approaches to scoring essays","volume":"14","author":"M Johnson","year":"2024","journal-title":"Sci Rep"},{"issue":"8","key":"pone.0354680.ref062","doi-asserted-by":"crossref","DOI":"10.1111\/lnc3.12432","article-title":"Five sources of bias in natural language processing","volume":"15","author":"D Hovy","year":"2021","journal-title":"Lang Linguist Compass"},{"key":"pone.0354680.ref063","unstructured":"Schaller NJ, Ding Y, Horbach A, Meyer J, Jansen T. In: Proceedings of the 19th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2024), 2024. 210\u201321. https:\/\/aclanthology.org\/2024.bea-1.18"},{"key":"pone.0354680.ref064","first-page":"54","article-title":"Integration of Automated Essay Scoring Models Using Item Response Theory","volume-title":"Lecture Notes in Computer Science","author":"I Aomi","year":"2021"},{"key":"pone.0354680.ref065","doi-asserted-by":"crossref","first-page":"21","DOI":"10.1016\/j.asw.2016.07.004","article-title":"A Many-Facet Rasch analysis comparing essay rater behavior on an academic English reading\/writing test used for two purposes","volume":"30","author":"S Goodwin","year":"2016","journal-title":"Assessing Writing"},{"key":"pone.0354680.ref066","first-page":"386","article-title":"Detecting and measuring rater effects using many-facet rasch measurement: Part I","volume":"4","author":"C Myford","year":"2003","journal-title":"Journal of Applied Measurement"},{"issue":"2","key":"pone.0354680.ref067","first-page":"189","article-title":"Detecting and measuring rater effects using many-facet Rasch measurement: Part II","volume":"5","author":"CM Myford","year":"2004","journal-title":"J Appl Meas"},{"key":"pone.0354680.ref068","doi-asserted-by":"crossref","unstructured":"Yang K, Rakovi\u0107 M, Li Y, Guan Q, Ga\u0161evi\u0107 D, Chen G. Unveiling the Tapestry of Automated Essay Scoring: A Comprehensive Investigation of Accuracy, Fairness, and Generalizability. 2024. https:\/\/arxiv.org\/abs\/2401.05655","DOI":"10.1609\/aaai.v38i20.30254"},{"key":"pone.0354680.ref069","unstructured":"Doewes A, Saxena A, Pei Y, Pechenizkiy M. In: Proceedings of the 15th International Conference on Educational Data Mining, 2022. 206\u201316."},{"issue":"2","key":"pone.0354680.ref070","doi-asserted-by":"crossref","first-page":"115","DOI":"10.1111\/j.1745-3984.1968.tb00613.x","article-title":"Test Bias: Prediction Of Grades Of Negro And White Students In Integrated Colleges","volume":"5","author":"TA Cleary","year":"1968","journal-title":"J Educational Measurement"},{"key":"pone.0354680.ref071","doi-asserted-by":"crossref","first-page":"142","DOI":"10.4324\/9781003278658-12","article-title":"Evaluating Fairness of Automated Scoring in Educational Measurement","volume-title":"Advancing Natural Language Processing in Educational Assessment","author":"MS Johnson","year":"2023"},{"issue":"9","key":"pone.0354680.ref072","doi-asserted-by":"crossref","first-page":"1263","DOI":"10.1109\/TKDE.2008.239","article-title":"Learning from Imbalanced Data","volume":"21","author":"He Haibo","year":"2009","journal-title":"IEEE Trans Knowl Data Eng"},{"key":"pone.0354680.ref073","unstructured":"Revelle W. Psych: Procedures for Psychological, Psychometric, and Personality Research. 2024."},{"issue":"3","key":"pone.0354680.ref074","doi-asserted-by":"crossref","first-page":"105","DOI":"10.15858\/engtea.78.3.202309.105","article-title":"Meta-analysis of inter-rater agreement and discrepancy between human and automated English essay scoring","volume":"78","author":"J Yun","year":"2023","journal-title":"English Teaching"},{"key":"pone.0354680.ref075","article-title":"The Duolingo English Test Responsible AI Standards","author":"J Burstein","year":"2025"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0354680","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T17:41:31Z","timestamp":1785519691000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0354680"}},"subtitle":[],"editor":[{"given":"Iftikhar Ahmed","family":"Khan","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,7,31]]},"references-count":75,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2026,7,31]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0354680","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,31]]}}}