@inproceedings{bediwy-etal-2026-seven,
title = "Seven Ways to Cut a Score: Mapping Essay-Scoring Predictions to Ordinal Ratings",
author = "Bediwy, Ahmed H. and
Bellows, Martha and
Lottridge, Sue",
editor = "Wilson, Joshua and
Ormerod, Christopher and
Beiting-Parrish, Magdalen",
booktitle = "Proceedings of the Artificial Intelligence in Measurement and Education Conference ({AIME}-Con): Full Papers",
month = oct,
year = "2026",
address = "Wyndham Grand Pittsburgh Downtown, Pittsburgh, Pennsylvania, United States",
publisher = "National Council on Measurement in Education (NCME)",
url = "https://aclanthology.org/2026.aimecon-main.41/",
pages = "372--379",
ISBN = "979-8-9983004-0-0",
abstract = "One approach to automated essay scoring re- lies on neural regression models that output continuous predicted scores, yet operational scores must be reported on a discrete, ordi- nal rubric scale to match human raters' scores. The mapping from continuous predictions to integer scores is governed by a small set of cutpoints, and the choice of cutpoint esti- mator materially changes both accuracy and the score distribution that examinees expe- rience. We introduce, formalize, and com- pare seven cutpoint methods{---}simple round- ing, exact-agreement optimization, balance op- timization, quadratic-weighted-kappa (QWK) maximization, maximum-likelihood Gaussian boundaries, a Bayesian ordered-prior estimator, and an ordinal cumulative link model{---}under a single notation. We evaluate all seven on seven prompts from the ASAP 2.0 dataset (24,728 essays), using a shared Longformer regres- sion backbone, and report agreement (quadratic weighted kappa), association (the Pearson cor- relation between integer machine and human scores), and distributional fidelity (the stan- dardized mean difference and the maximum score-point distribution gap). Our results ex- pose a consistent accuracy{--}calibration trade- off: QWK maximization and balance optimiza- tion tie for the highest agreement, balance opti- mization achieves the smallest score-point dis- tribution gap, the ordinal link model achieves the smallest mean bias, the generative Gaussian and Bayesian estimators recover score means but distort the distribution, and the ordinal link model is a strong all-rounder."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="bediwy-etal-2026-seven">
<titleInfo>
<title>Seven Ways to Cut a Score: Mapping Essay-Scoring Predictions to Ordinal Ratings</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ahmed</namePart>
<namePart type="given">H</namePart>
<namePart type="family">Bediwy</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Martha</namePart>
<namePart type="family">Bellows</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sue</namePart>
<namePart type="family">Lottridge</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Artificial Intelligence in Measurement and Education Conference (AIME-Con): Full Papers</title>
</titleInfo>
<name type="personal">
<namePart type="given">Joshua</namePart>
<namePart type="family">Wilson</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christopher</namePart>
<namePart type="family">Ormerod</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Magdalen</namePart>
<namePart type="family">Beiting-Parrish</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>National Council on Measurement in Education (NCME)</publisher>
<place>
<placeTerm type="text">Wyndham Grand Pittsburgh Downtown, Pittsburgh, Pennsylvania, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-9983004-0-0</identifier>
</relatedItem>
<abstract>One approach to automated essay scoring re- lies on neural regression models that output continuous predicted scores, yet operational scores must be reported on a discrete, ordi- nal rubric scale to match human raters’ scores. The mapping from continuous predictions to integer scores is governed by a small set of cutpoints, and the choice of cutpoint esti- mator materially changes both accuracy and the score distribution that examinees expe- rience. We introduce, formalize, and com- pare seven cutpoint methods—simple round- ing, exact-agreement optimization, balance op- timization, quadratic-weighted-kappa (QWK) maximization, maximum-likelihood Gaussian boundaries, a Bayesian ordered-prior estimator, and an ordinal cumulative link model—under a single notation. We evaluate all seven on seven prompts from the ASAP 2.0 dataset (24,728 essays), using a shared Longformer regres- sion backbone, and report agreement (quadratic weighted kappa), association (the Pearson cor- relation between integer machine and human scores), and distributional fidelity (the stan- dardized mean difference and the maximum score-point distribution gap). Our results ex- pose a consistent accuracy–calibration trade- off: QWK maximization and balance optimiza- tion tie for the highest agreement, balance opti- mization achieves the smallest score-point dis- tribution gap, the ordinal link model achieves the smallest mean bias, the generative Gaussian and Bayesian estimators recover score means but distort the distribution, and the ordinal link model is a strong all-rounder.</abstract>
<identifier type="citekey">bediwy-etal-2026-seven</identifier>
<location>
<url>https://aclanthology.org/2026.aimecon-main.41/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>372</start>
<end>379</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Seven Ways to Cut a Score: Mapping Essay-Scoring Predictions to Ordinal Ratings
%A Bediwy, Ahmed H.
%A Bellows, Martha
%A Lottridge, Sue
%Y Wilson, Joshua
%Y Ormerod, Christopher
%Y Beiting-Parrish, Magdalen
%S Proceedings of the Artificial Intelligence in Measurement and Education Conference (AIME-Con): Full Papers
%D 2026
%8 October
%I National Council on Measurement in Education (NCME)
%C Wyndham Grand Pittsburgh Downtown, Pittsburgh, Pennsylvania, United States
%@ 979-8-9983004-0-0
%F bediwy-etal-2026-seven
%X One approach to automated essay scoring re- lies on neural regression models that output continuous predicted scores, yet operational scores must be reported on a discrete, ordi- nal rubric scale to match human raters’ scores. The mapping from continuous predictions to integer scores is governed by a small set of cutpoints, and the choice of cutpoint esti- mator materially changes both accuracy and the score distribution that examinees expe- rience. We introduce, formalize, and com- pare seven cutpoint methods—simple round- ing, exact-agreement optimization, balance op- timization, quadratic-weighted-kappa (QWK) maximization, maximum-likelihood Gaussian boundaries, a Bayesian ordered-prior estimator, and an ordinal cumulative link model—under a single notation. We evaluate all seven on seven prompts from the ASAP 2.0 dataset (24,728 essays), using a shared Longformer regres- sion backbone, and report agreement (quadratic weighted kappa), association (the Pearson cor- relation between integer machine and human scores), and distributional fidelity (the stan- dardized mean difference and the maximum score-point distribution gap). Our results ex- pose a consistent accuracy–calibration trade- off: QWK maximization and balance optimiza- tion tie for the highest agreement, balance opti- mization achieves the smallest score-point dis- tribution gap, the ordinal link model achieves the smallest mean bias, the generative Gaussian and Bayesian estimators recover score means but distort the distribution, and the ordinal link model is a strong all-rounder.
%U https://aclanthology.org/2026.aimecon-main.41/
%P 372-379
Markdown (Informal)
[Seven Ways to Cut a Score: Mapping Essay-Scoring Predictions to Ordinal Ratings](https://aclanthology.org/2026.aimecon-main.41/) (Bediwy et al., AIME-Con 2026)
ACL