Eedi Misconceptions Graph: 8,117 Student Math Misconceptions Mapped to 2,258 Curriculum Constructs (Eedi)
- Provider
- Eedi
- License
- CC-BY-4.0 (source)
- Size
- 12 files
- Format
- CSV (in sampled files)
- Stored at
- Hugging Face (https://huggingface.co/datasets/Eedi/Eedi-Misconceptions-Graph)
- Access
- Free and public. No account with the data provider is needed.
- Category
- Mathematics
- Source
- https://www.eedi.com
About this dataset
A map of mathematical misconceptions held by school students and of the curriculum skills (constructs) in which they appear, released by the education company Eedi. Students on the Eedi platform answer multiple-choice Diagnostic Questions in which each wrong answer (distractor) is tagged with the misconception it reveals, for example a systematic error when ordering fractions. Aggregating those taggings gives misconception-map.csv, a weighted bipartite edge list with 19,294 rows covering 8,117 misconceptions. The edge weight, PercentageMisconceptionExistsInThisConstruct, is the share of a misconception's taggings that fall in each construct. construct-map.csv catalogs 2,258 constructs in a subject, topic and sub-topic hierarchy (4 subjects, 50 topics, 236 sub-topics: Number, Geometry and Measure, Algebra, Data and Statistics). The question texts themselves are not included. Education researchers use it to study how errors spread across topics, to design diagnostic assessments, and to ground tutoring models. Load the CSVs with pandas; scripts/ converts them to graph JSONL for Neo4j. The data is CC BY 4.0 and the scripts are MIT. Attribute "Eedi Misconceptions Graph v1.0" by Eedi.
How to load this dataset in Python
import pandas as pd
df = pd.read_csv("hf://datasets/Eedi/Eedi-Misconceptions-Graph/data/construct-map.csv")
df.head()Data structure
File types in sampled files: CSV
Files
Example file: data/construct-map.csv
Analyze this dataset without downloading it
On Scigantic, Eedi Misconceptions Graph: 8,117 Student Math Misconceptions Mapped to 2,258 Curriculum Constructs (Eedi) opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.
This page is also available as Markdown and as JSON.