Breadcrumb
- Home
- AI Hub For Education
- Research Study Repository
- Impact – Quasi–experimental
Search and Filter
The Repository
Impact – Quasi–experimental
Research synthesis is AI-generated, human reviewed. Updated 11/2026.
Displaying 1 - 30 of 102
The Effects Of Course-Integrated AI Tutoring On Student Performance And Engagement: A Randomized University Trial
Jing Liu, Tracy Sweet, Marty Haoyuan Chen, Jonathan Engelberg, Megan C. Masters, Martyn Clark, Axel Persaud, Alia Lancaster, Jeffrey Hollingsworth, Jennifer King Rice. (10/2026). EdWorking Papers.
Among sections of the same course, access to the AI tutor reduced final grades by 0.37 standard deviations and learning management system participation by 0.90 standard deviations.
Teacher-Mediated Multimodal Platform And Math Achievement: Evidence From Goblins' Yearlong District Implementation
Anika Alam, Nicolas Dominguez-Mel, Jing Liu. (10/2026). EdWorking Papers.
Using the Goblins platform is associated with a 0.11 standard deviation increase in MAP math scores within the district.
Little Impact Of ChatGPT On High School Test Scores
Nick Huntington-Klein. (09/2026). arXiv.
Higher educational ChatGPT use had an insignificant effect on high school test scores that was close to zero, bounded below roughly 0.04 standard deviations.
Dose, Not Access: Digital Reading Tutor Use And Early Reading Growth
Nicholas A. Gage. (09/2026). EdArXiv.
High-dose users averaging at least 15 minutes weekly outperformed near-zero users on end-of-year DIBELS reading composite scores by a medium effect size.
AI-Written Admissions Essays Are Widespread But Penalized
Calvin Isley, Johann Gaebler, Sharad Goel. (09/2026). arXiv.
By 2025, 56% of applicants submitted at least one primarily AI-generated essay—reaching 69.3% among international applicants—despite an explicit prohibition against AI use.
Generative AI And Extended Reality In Collaborative Architectural Design Education: An Exploratory Studio Study
Yao Xiao, Max Chen, Yichen Li, Nathaniel Powers, Maxwell Wiesenfeld, Gillian Smith, Soroush Farzin, Shichao Liu. (09/2026). arXiv.
Students using the GenARch pipeline experienced larger relative declines in design self-efficacy confidence and outcome expectancy, alongside a significant relative gain in perceived conflict management.
Beyond Access: Guided LLM Scaffolding For Independent Learning In Undergraduate Statistics
Mohammad Amanloua, Yasaman Amou-Jafaria, Mehrad Liviana, Fatemeh Boloukazari a, Fereshte Bagheria, Behnam Bahrakb. (08/2026). arXiv.
Guided AI use was associated with higher scores on unassisted quizzes, peaking on the final quiz where guided students outscored unguided peers by 16.42 points.
Making AI-Generated Feedback Matter: A Large-Scale Study Of Feedback Workflows And Student Enactment
Omar Alsaiari, Nilufar Baghaei, Jason M. Lodge, Naomi Winstone, Dragan Gašević, Hassan Khosravi. (08/2026). arXiv.
Enacted Feedback was associated with significantly higher feedback uptake (estimated probability of 26.2%) compared with Directed Feedback (14.1%) and Self-Directed Feedback (0.1%).
Methodologies For Improving The Quality Of AI Tutoring In K-12 Education
Tushar Udeshi, Anna Khazenzon, Kabir Khan, Nick Breen, RJ Corwin, Chris DiGiano, Kodi Weatherholtz, Marek Zaluski. (08/2026). arXiv.
Migrating the main Exercise AI tutor from GPT-4o to GPT-4.1 reduced giving away final answers by 96.66% and increased behavioral engagement by 9.93%.
Structured AI Demonstrations And Student LLM Use In Engineering Mechanics: Study Design And Preliminary Results
Shuang Geng, Helen Lallos-Harrell, Jiya Ashar, Thomas J. McKenna, Annwesa Dasgupta, Caleb Farny, Emma Lejeune. (07/2026). arXiv.
Undergraduate use of artificial intelligence for engineering mechanics rose from 37% weekly use at baseline to 51% by the end of the semester.
Generative AI Availability, Grades, And Student Satisfaction At A Large University
James M. Zumel Dumlao, Meng Wang, Zhonghan Xie, Junyao Hu, Ivan Bar, George Chaney III, Henry Gold, Misha Teplitskiy. (07/2026). arXiv.
Effects of ChatGPT's release on student grades were small and non-robust, ranging from 0.010 to 0.071 grade points depending on how pandemic trends were modeled.
Experiential Versus Instructional Approaches For Eliciting Metacognitive Awareness In AI-Assisted Learning: A Short-Term Longitudinal Study
Pau Benazet i Montobbio, Janne Rotter, Davinia Hernandez-Leo. (07/2026). arXiv.
Experiential learning produced significantly higher immediate gains in AI engagement and knowledge of cognition than lecture instruction.
Teaching AI Interactively: An Experience Report In Higher Education
Jennifer M. Reddig, Scott Moon, Kaitlyn Crutcher, Christopher J. MacLellan. (07/2026). arXiv.
Summer students had 6.8 times higher odds of reporting higher attendance than students in the traditional spring lecture offering.
Faster Completion, Less Learning: Generative AI Reduced Study Time On Math Problems And The Knowledge They Build
Sina Rismanchian, Hasan Uzun, Jeffrey Matayoshi, Eric Cosyn, Eyad Kurd-Misto. (07/2026). arXiv.
Learning time on AI-susceptible math word problems dropped cumulatively by 31.3% for high schoolers and 26.9% for college students after ChatGPT's release, with no change in Grade 5.
The Generative AI Learning Penalty: Evidence from Chinese Secondary Education
David Stromberg, Victor Lei, Yanhui Wu. (07/2026). SSRN.
AI adoption raised homework scores by 18% and cut completion time by 30%, but lowered monthly exam scores by 20% within six months.
Teaching Students To Question The Machine: Short-Term Effects Of An AI Literacy Workshop On Middle-School Students' Regulation Of LLM Interaction
Olivier Clerc, Rania Abdelghani, Chloé Desvaux, Eliott Poisson, Pierre-Yves Oudeyer, Helene Sauzeon. (06/2026). arXiv.
Trained students demonstrated better interaction behaviors, accepting underspecified prompts less often (51.5% vs. 66.7%), asking more follow-up questions (59.2% vs. 27.9%), and judging response correctness more accurately.
LecturaRaagents: A Multi-Agent Framework For Adaptive Personalized AI-Assisted Learning And Embodied Teaching
Jaward Sesay, Yue Yu, Siwei Dong, Guangyao Chen, Börje F. Karlsson. (06/2026). arXiv.
The new multi-agent system consistently outperformed existing approaches in lecture content quality, embodied teaching quality, assessment, and personalization across high school, undergraduate, and graduate courses.
The Creative Link Between Words and Ideas is Weakening in the AI Era
Kibum Moon, Kostadin Kushlev, Andrew Bank, Benjamin Lira, Indre Viskontas, James C. Kaufman, Dan Richard Johnson, Angela Lee Duckworth, Adam E. Green. (06/2026). PsyArXIV.
Comparing essays before and after ChatGPT's release, word diversity increased while conceptual diversity within and between essays decreased.
Detective Scaffolding For Within-Session Reasoning Development: A Three-Phase Framework Evaluated In Polymer Engineering And Pre-University Outreach
Haolin Feng, Holly Barrett, Xinru Deng, Dimitrios G. Papageorgiou, Yiwei Sun. (06/2026). arXiv.
Undergraduate teams completely shifted from an initial misconception (71% selecting temperature) to 100% identifying humidity as a contributing factor across four staged polls.
Who Benefits from LLM-Generated Learning Support Messages? Causal Evidence on Treatment Effect Heterogeneity
Aylin Ozturk, Gati Aher, Tom Mitchell, Mustafa Kemal Birgin, Huseyin Kayhan. (06/2026). Artificial Intelligence in Education.
Assignment to receive AI-generated support messages improved overall course scores by 0.60 points and final exam participation by 1.9 percentage points.
Combating Harms Of Generative AI In Cs1 With Code Review Interviews And A Flipped Classroom
Peter Fowles, Erik Falor, Sulove Bhattarai, John Edwards, Seth Poulsen. (05/2026). arXiv.
Exam performance showed no statistically significant differences across cohorts, with Fall 2025 students scoring about two percent higher on average than prior cohorts on Exam 1.
Gender Differences In AI Literacy Workshop Outcomes And Deepfake Engagement
Jake Renzella, Christian Bergh, Natasha Banks, Alexandra Vassar. (05/2026). arXiv.
Before the workshop, male students reported significantly higher career interest across engineering, computer science, and AI, while female students were more likely to use AI for schoolwork.
Teaching with Gemini: Measuring the impact of Guided Learning on student mathematics progress in Sierra Leone
LearnLM Team, Google & Fab AI. (05/2026). Google.
Gemini Guided Learning significantly improved student mathematics learning outcomes, with an intent-to-treat effect of +0.258 standard deviations.
Little Impact Of ChatGPT Availability On High School Student Test Score Performance
Nick Huntington-Klein. (05/2026). arXiv.
Increased educational ChatGPT exposure has no statistically significant effect on high school test scores, changing scores by an estimated -0.0096 standard deviations in the baseline model.
Code Comprehension With Github Copilot: Performance Gains, Comprehension Trade-Offs, And Behavioral Predictors In Brownfield Programming
Yunhan Qiao, Md Istiak Hossain Shihab, Summit Haque, Christopher Hundhausen. (05/2026). arXiv.
GitHub Copilot cut completion time by 50.7 percent and increased passed tests by 71.4 percent, but produced no corresponding improvement in overall code comprehension.
Lata: A Drop-In, Ferpa-Compliant Local-LLM Autograder For Upper-Division STEM Coursework
Jesse A. Rodríguez. (05/2026). arXiv.
The local autograder processed submissions in 1 to 3 minutes each, with an instructor-confirmed grading-error rate of roughly 0.02% to 0.04% per rubric line item.
Maic-Ui: Making Interactive Courseware With Generative Ui
Shangqing Tu, Yanjia Li, Keyu Chen, Sichen Zhang, Jifan Yu, Daniel Zhang-Li, Lei Hou, Juanzi Li, Yu Zhang, Huiqin Liu. (04/2026). arXiv.
In a high school deployment, the pilot class achieved 9.21-point average gains in STEM subjects compared to -2.32 points across other classes.
Al-Assisted Code Review As A Scaffold For Code Quality And Self-Regulated Learning: An Experience Report
Eduardo Araujo Oliveira, Michael Fu, Patanamon Thongtanunam, Sonsoles López-Pernas, Mohammed Saqr. (04/2026). arXiv.
Across both cohorts, student responsiveness was stable, with 32 percent of reviewed pull requests in 2023 and 33 percent in 2024 followed by subsequent code commits.
Artificial Intelligence As A Real Game To Enlighten Science Education For Disabled Students In Rural New Mexico
Nelson Uloma Egondu, Gil Gallegos. (04/2026). arXiv.
Following an 8-week AI intervention, students improved science concept retention by 32%, laboratory performance by 27%, and classroom engagement by 42%.
Argumath: AI-Simulated Environment For Pre-Service Teacher Training In Orchestrating Classroom Mathematics Argumentation
Jiwon Chun, Yuling Zhuang, Armanto Sutedjo, Colin Xu, Rong Ren, Meng Xia. (04/2026). arXiv.
Pre-service teachers using ArguMath elicited more student responses (89 vs. 76) and asked more framework-coded questions (68 vs. 49) than when using the baseline system.
