README.md

metadata

license: cc0-1.0
language:
  - en
tags:
  - synthetic
  - template-generated
  - code
  - computerscience
  - education
  - k-12
  - k12
  - technology
  - tech
pretty_name: K-12 Computer Science Standards (generated)
size_categories:
  - n<1K

K-12 Computer Science Standards

696 generated learning-objective records organized around K-12 computer science concept areas, including computing systems, networks, data, algorithms, programming, AI/ML, cybersecurity, data science, and robotics.

How this was built (read this first)

These records are programmatically generated, not transcribed from official standards documents. A generator took a standards taxonomy - codes, grade levels, domains, and similar structured fields - and expanded it into instruction/output pairs using a fixed set of sentence templates.

Two consequences:

1. The text is not official standard language. Where a field looks like a standard's wording, it is template output built around the standard's code and domain, not the text the issuing body published.
2. The text repeats heavily. Of 556 training rows only 266 learning objectives are distinct (47.8%), and the phrasing is formulaic - "Students will understand and apply {topic} in computing contexts" recurs throughout. The grade-band distribution is exactly 174 records per band, which is a generator artifact.

Field values that are drawn from the source taxonomy - standard codes, grade levels, domain and framework names - are reliable. The generated prose around them is not.

Loading

from datasets import load_dataset

ds = load_dataset("robworks-software/k12-computer-science-standards")

Splits

Split	Rows
train	556
test	140
total	696

Concept distribution (training split)

Concept	Rows
Programming	186
Data Science	53
Robotics	51
Cybersecurity	49
Algorithms and Programming	49
Artificial Intelligence	47
Data and Analysis	38
Networks and the Internet	29

Limitations

• Not CSTA or ISTE standard text. The CSTA K-12 CS Standards define a specific set of standards with specific wording. This dataset is generated around CSTA's concept structure; it does not reproduce CSTA standards and should not be cited as CSTA content. CSTA standards are available from csteachers.org.
• 47.8% distinct objectives - see above.
• AI, cybersecurity, data science and robotics are not CSTA concept areas. Those four categories (274 of 696 records, nearly 40%) were added by this project. They do not correspond to any published framework.
• Tool and platform references are opinions, and some are inappropriate as written. The concept areas suggest specific tooling by grade band; a prior version of this card listed penetration-testing tooling for grades 9-12 as though it were standards content. Treat all tool suggestions as unreviewed authoring, not curriculum guidance.
• No expert review. A previous version of this card claimed "Expert Review". That did not happen.
• US-centric and reflects a 2024 snapshot of the field.

Source taxonomy

Concept structure informed by the CSTA K-12 Computer Science Standards and ISTE computational thinking competencies. Generated text is original output of this project and is not endorsed by CSTA or ISTE.

License

CC0-1.0 for this compilation and its generated text.

Citation

@dataset{k12_cs_standards,
  title  = {K-12 Computer Science Standards},
  author = {Robson, Ryan and Robworks Software},
  year   = {2025},
  publisher = {Hugging Face},
  note   = {Programmatically generated around CSTA/ISTE concept structure; not official standard text},
  url    = {https://huggingface.co/datasets/robworks-software/k12-computer-science-standards}
}