Subtopic Deep Dive

Automated Essay Scoring
Research Guide

What is Automated Essay Scoring?

Automated Essay Scoring (AES) uses natural language processing to automatically grade written essays by evaluating traits such as coherence, argumentation, grammar, and content relevance with accuracy comparable to human raters.

AES systems employ machine learning models, including early statistical methods and modern transformer architectures, to score essays on rubrics matching human judgments. Research focuses on improving reliability across diverse topics and languages while addressing biases. Over 10 papers in the provided list discuss AI applications in education, with generative AI papers like Alier et al. (2024) cited 120 times exploring related assessment disruptions.

11
Curated Papers
3
Key Challenges

Why It Matters

AES scales writing assessment for large classes, providing instant feedback to improve student writing skills in K-12 and higher education (Chan and Colloton, 2024). It reduces teacher workload in formative evaluation, enabling personalized instruction amid growing enrollment. In GenAI contexts, AES detects AI-generated content to maintain academic integrity (Alier et al., 2024; Klopfer et al., 2024). Systems support distance learning by automating grading of programming essays and key competence development (Shokaluk et al., 2020).

Key Research Challenges

Bias in Scoring Models

AES models exhibit biases against non-native English speakers or underrepresented topics, reducing fairness (Selwyn, 2024). Validation against diverse human raters remains inconsistent. Chan and Colloton (2024) highlight equity issues in GenAI grading applications.

GenAI Detection Reliability

Distinguishing student-written from GenAI-generated essays challenges AES validity post-ChatGPT (Alier et al., 2024). Current detectors fail on edited AI outputs. Klopfer et al. (2024) note risks in K-12 assessments.

Scalability Across Languages

Most AES tools perform poorly on non-English essays despite multilingual needs (Chisega-Negrilă, 2023). Training data scarcity limits generalization. Shokaluk et al. (2020) discuss competence assessment in distance learning.

Essential Papers

1.

Generative Artificial Intelligence in Education: From Deceptive to Disruptive.

Marc Alier, Francisco José García‐Peñalvo, Jorge D. Camba · 2024 · International Journal of Interactive Multimedia and Artificial Intelligence · 120 citations

Generative Artificial Intelligence (GenAI) has emerged as a promising technology that can create original content, such as text, images, and sound. The use of GenAI in educational settings is becom...

2.

Generative AI in Higher Education

Cecilia Ka Yuk Chan, Tom Colloton · 2024 · 105 citations

Chan and Colloton’s book is one of the first to provide a comprehensive examination of the use and impact of ChatGPT and Generative AI (GenAI) in higher education.
\n
\nSince November 2022,...

3.

On the Limits of Artificial Intelligence (AI) in Education

Neil Selwyn · 2024 · Nordisk tidsskrift for pedagogikk og kritikk · 77 citations

The recent hyperbole around artificial intelligence (AI) has impacted on our ability to properly consider the lasting educational implications of this technology. This paper outlines a number criti...

4.

Technologies of distance learning for programming basics on the principles of integrated development of key competences

Світлана Вікторівна Шокалюк, Yelyzaveta Yu. Bohunenko, Iryna Lovianova et al. · 2020 · CTE Workshop Proceedings · 47 citations

In the era of the fourth industrial revolution – Industry 4.0 – developing key competences (digital, multilingual and mathematical competences in particular) is of paramount importance. The purpose...

5.

Artificial Intelligence and Teachers’ New Ethical Obligations

Catherine Adams, Patti Pente, Gillian Lemermeyer et al. · 2022 · The International Review of Information Ethics · 37 citations

Largely thought to be immune from automation, the teaching profession is now being challenged on multiple fronts by new digital infrastructures and smart software that automate pedagogical decision...

6.

Artificial Intelligence and K-12 Education: Possibilities, Pedagogies and Risks

Joseph Mintz, W. Holmes, Leping Liu et al. · 2023 · Computers in the Schools · 37 citations

7.

Generative AI and K-12 Education: An MIT Perspective

Eric Klopfer, Justin Reich, Hal Abelson et al. · 2024 · 20 citations

In November of 2022, a Silicon Valley company launched an invention that could complete students' homework for them. Available only to subscribers at first, by the spring of 2023 OpenAI's ChatGPT-3...

Reading Guide

Foundational Papers

Davis (1987) provides early social contract view on math education assessment, foundational for AES equity discussions despite low citations.

Recent Advances

Alier et al. (2024) and Chan and Colloton (2024) detail GenAI disruptions to essay scoring; Klopfer et al. (2024) covers K-12 specifics.

Core Methods

NLP trait analysis via transformers for coherence/argumentation; statistical validation (kappa); GenAI for synthetic data augmentation (Alier et al., 2024).

How PapersFlow Helps You Research Automated Essay Scoring

Discover & Search

PapersFlow's Research Agent uses searchPapers and exaSearch to find AES-related papers like 'Generative Artificial Intelligence in Education' by Alier et al. (2024), then citationGraph reveals connections to Chan and Colloton (2024) for GenAI impacts, and findSimilarPapers uncovers K-12 applications from Klopfer et al. (2024).

Analyze & Verify

Analysis Agent applies readPaperContent to extract AES validation metrics from Selwyn (2024), verifies claims with verifyResponse (CoVe) against human rater correlations, and uses runPythonAnalysis for statistical verification of scoring agreement (e.g., Cohen's kappa via pandas/NumPy); GRADE grading assesses evidence strength in bias discussions.

Synthesize & Write

Synthesis Agent detects gaps in GenAI detection via contradiction flagging across Alier et al. (2024) and Klopfer et al. (2024); Writing Agent employs latexEditText for rubric tables, latexSyncCitations to link Shokaluk et al. (2020), and latexCompile for polished reports with exportMermaid diagrams of model architectures.

Use Cases

"Compare Cohen's kappa scores for AES bias in GenAI essays across recent edAI papers."

Research Agent → searchPapers('AES bias GenAI') → Analysis Agent → runPythonAnalysis (extract kappa from Alier 2024, Selwyn 2024 via pandas) → GRADE grading → CSV export of statistical comparisons.

"Draft LaTeX section on AES challenges with citations from provided edAI papers."

Synthesis Agent → gap detection (bias, GenAI) → Writing Agent → latexEditText (add rubric) → latexSyncCitations (Alier 2024, Chan 2024) → latexCompile → PDF output with compiled bibliography.

"Find GitHub repos with AES model code from education AI papers."

Research Agent → searchPapers('AES code education') → Code Discovery → paperExtractUrls → paperFindGithubRepo (e.g., from Chisega-Negrilă 2023 links) → githubRepoInspect → summary of trainable models.

Automated Workflows

Deep Research workflow conducts systematic review of 50+ edAI papers on AES, chaining searchPapers → citationGraph → structured report with GRADE scores. DeepScan applies 7-step analysis with CoVe checkpoints to verify GenAI detection claims in Alier et al. (2024). Theorizer generates hypotheses on AES fairness from Shokaluk et al. (2020) and Selwyn (2024) literature.

Frequently Asked Questions

What is Automated Essay Scoring?

AES applies NLP to grade essays on traits like coherence and argumentation, matching human rater scores using models from statistical to transformer-based.

What methods dominate AES research?

Methods include machine learning for trait scoring and recent GenAI integration for feedback; transformers handle argumentation (Alier et al., 2024).

What are key papers on AES in edAI?

Alier et al. (2024, 120 citations) covers GenAI disruptions; Chan and Colloton (2024, 105 citations) examines higher ed impacts; Klopfer et al. (2024) addresses K-12 risks.

What open problems exist in AES?

Challenges include bias mitigation, GenAI detection accuracy, and multilingual scalability (Selwyn, 2024; Chisega-Negrilă, 2023).

Research Artificial Intelligence in Education with AI

PapersFlow provides specialized AI tools for Computer Science researchers. Here are the most relevant for this topic:

See how researchers in Computer Science & AI use PapersFlow

Field-specific workflows, example queries, and use cases.

Computer Science & AI Guide

Start Researching Automated Essay Scoring with AI

Search 474M+ papers, run AI-powered literature reviews, and write with integrated citations — all in one workspace.

See how PapersFlow works for Computer Science researchers