Loading...
Loading...
Student performance, MOOC logs, knowledge tracing, standardized tests, learning analytics
13,362 datasets
NOAA's Southeast Area Monitoring and Assessment Program (SEAMAP) collected temperature, salinity, and dissolved oxygen measurements via CTD and bottle casts from the NOAA Ship OREGON II in the Gulf of Mexico. The dataset represents a State/Federal/university collaborative effort for fishery-independent data collection and spans the period from 2001 to 2012. It is managed by the National Oceanic and Atmospheric Administration, Department of Commerce.
Shanghai Jiao Tong University's annual ranking of global universities from 2003 to 2024. The dataset likely contains scores and ranks based on academic and research performance indicators. Its long time series allows for analysis of institutional performance trends over more than two decades.
Kaggle hosts this dataset of speech features and pronunciation assessment records. The data is intended for defect classification in oral English pronunciation. The author, organization, and specific scale of the collection are not provided.
Kaggle dataset titled 'dropoutlens_ai' likely concerning student dropout prediction. The dataset's specific content, size, and origin are unverified from the provided metadata. Its columns and structure must be inspected after download.
An AusGeo News article outlines the geological and petroleum prospectivity assessment for the Capel and Faust basins. The assessment was conducted by Geoscience Australia's Remote Eastern Frontiers project from 2006 to 2010. It details the regional setting, data acquisition methods, assessment methodology, and study findings.
The 2015 annual report on educational performance for London, published by the Greater London Authority. It is the third such report in a series, with underlying data available via a provided web link. The dataset was last updated in the platform's metadata on 2026-03-25.
The National Center for Education Statistics (NCES) collects and reports statistics on the condition of education in the United States and other nations. This dataset, authored by Michael F. Middaugh, focuses on instructional costs and productivity in higher education, likely containing expenditure and productivity metrics for academic departments or institutions.
A study examining teacher sorting within schools, comparing teachers in the same grade and school. The research finds associations between teacher experience, race, gender, and the assignment of students with lower prior achievement, more behavioral problems, and lower attendance. The dataset likely contains variables related to teacher demographics, student prior performance, and class assignments.
William Sanders developed the Tennessee Value Added Assessment System (TVAAS) to evaluate teacher influence on student learning. This paper by Haggai Kupermintz examines the validity of TVAAS teacher effectiveness measures, analyzing claims about their ability to capture unique teacher contributions and guide instructional practice. The system uses a mixed-effects model applied to longitudinal standardized test score data across several subject areas.
64 children's social competence ratings collected during a randomized controlled trial of an 11-week equine facilitated learning program. Parents provided ratings at pretest and posttest for both an experimental group and a waitlisted control group. The dataset was created by researcher Patricia Pendry to study causal effects of equine interventions.
A six-year longitudinal study of 158 freshmen tracks cognitive and affective factors related to timely degree completion and cumulative GPA. The dataset likely contains Scholastic Achievement Test (SAT) scores, metacognitive skills, locus of control, interpersonal support, self-efficacy, and action behaviors. It was created by researcher Cathy W. Hall.
A research paper presents findings from a study investigating how learning communities were created and sustained in 17 charter schools in the United States. The study examined school missions, instructional programs, accountability systems, and leadership, identifying four critical building blocks and three enabling conditions. The paper was authored by Priscilla Wohlstetter and presented at the American Educational Research Association meeting in March 1997.
A 1990 analysis of students who were in eighth grade in 1988 but not enrolled in school by 1990. The study found differences in dropout reasons and plans to resume education based on race-ethnicity and gender. It was authored by Will J. Jordan and references data from the National Education Longitudinal Study of 1988.
A field operational test of prototype integrated crash warning systems conducted over a 6-week period for 108 light-vehicle drivers and a 10-month period for 18 heavy-truck drivers. The report, authored by James R. Sayer, presents findings from the University of Michigan Transportation Research Institute on driver behavior and acceptance. Data captured includes the driving environment, driver behavior, warning system activity, and vehicle kinematics.
Paperswithcode hosts a collection of academic papers analyzing K-16 education reform efforts across seven U.S. states. The work, authored by Michael W. Kirst and other researchers, examines systemic barriers and reform strategies for improving student transitions from high school to college. The raw description lists chapters dedicated to California, Texas, Illinois, Oregon, Georgia, Maryland, and the role of community colleges.
Oceanographic data from the Gulf of Mexico collected by Texas A&M University aboard the R/V Gyre during a one-week cruise in November 1989. The dataset includes cloud amount/frequency, nitrate concentrations, and other physical profile measurements. Data were submitted by Dr. David Murphy and processed by the National Oceanographic Data Center.
CAD-S is the first openly available dataset for resume credibility assessment using NLP. It supports supervised learning for detecting inconsistencies between claimed skills and supporting evidence within resumes. The dataset was created by aselasperera and was last updated in March 2026.
Titanic passenger data is a canonical benchmark for binary classification tasks in machine learning education. The dataset is published on Kaggle, a platform for data science competitions and projects. Its exact size, features, and provenance are unspecified in the provided metadata.
Education Autolabel is a dataset published on HuggingFace by author shyuni. The dataset likely contains labeled data for educational applications, inferred from its title. Its last update was recorded on 2026-05-01 12:58:12.
1,739,249 tokens of text data generated by the Qwen3.6-plus model for knowledge distillation. The dataset covers topics including coding, mathematics, finance, medicine, and economics, with a maximum sequence length of 6,500 tokens per row. It was created by author 'ansulev' and last updated on April 8, —.