educational-assessment

Classification model to predict student performance in the Saber Pro exams in Colombia. This repository includes exploratory data analysis, data preprocessing, and machine learning models. Ideal for educational data scientists and researchers interested in academic performance prediction.

ENEM 2020 — Efeito da posição do item na probabilidade de acerto Análise estatística e preditiva do efeito da posição das questões de Matemática no ENEM 2020, combinando regressão logística, bootstrap e modelos de Machine Learning com foco em inferência e desempenho preditivo.

Educational Assesement using LLMs

Educational clustering analysis using K-Means and statistical methods based on a peer-reviewed scientific publication.

Characteristic curves of MCQ alternatives

Automated metadata extraction pipeline for International Large-Scale Assessment (ILSA) documents using RAG-based LLM architecture. 99.2% classification accuracy across 1,622 studies.

🧠 psychometrics-and-assessment-program · 🧪 5 instrumentos · 📝 344 ítems · 📐 81 escalas · ⚙️ scoring auditable · 🎓 currículo: 40 clases (5 escritas) · ⚖️ licencias trazables · 🐍 Python 3.11–3.13 · ✅ 57 tests

Code, per-item results and figures for a study dissociating prompt quality from response compliance in automated prompt-engineering assessment. Four-agent MATLAB evaluator on a locally hosted Qwen 2.5-7B judge, over 498 prompts from IFEval, LMSYS-Chat-1M and WildChat.

Web app for managing Assessment Content Packages (ACPs) for educational assessments.

Research code analyzing the relationship between word predictability and other indicators of lexical proficiency in language learner writing.

Context-aware formative feedback evaluator with multidimensional rubrics, evidence boundaries, error analysis, and human-rater validation.

This project analysed the competency level of test construction among teachers in senior secondary schools in Edo State, Nigeria. Using Python, the analysis leveraged libraries like scipy.stats (ttest_ind, levene), pandas, matplotlib, numpy, and seaborn to assess teachers' proficiency and identify areas for improvement.

AI-assisted analytic assessment that keeps the teacher's judgement central. Two Model Context Protocol servers for Claude Desktop; the assessment logic lives in editable methodology documents.