brightkey/uni-rankings-2026

BrightKey Independent University Rankings Dataset (2026)

199 universities × 15 countries × 6 dimensions, evaluated independently. No payments from institutions accepted. Public data only.

This is the open release of the BrightKey university rankings — an independent alternative to QS, THE, and Shanghai rankings. Released under CC BY 4.0.

• Live site: https://brightkey.co/en/rankings/methodology
• GitHub repo: https://github.com/arthurb2l/brightkey-university-dataset
• Wikidata: Q140002954

Why this exists

Commercial university rankings depend on:

• Reputation surveys that measure brand recognition, not educational quality
• Money — universities pay for inclusion, advertising, premium analysis
• Black-box weighting reverse-engineered to produce desired results

This dataset is different:

• No payments from any university, ever
• Public data only — anyone can verify
• Tier profiles (S/A/B/C/D), not misleading ordinal ranks
• Missing data is flagged, never penalized
• Post-study work visa pathways included (no commercial ranking does this)

Dataset structure

Three CSV configs:

universities (default config)

199 rows × 23 columns. One row per university.

Column	Type	Description
slug	string	URL-safe unique identifier
name	string	Official institution name
country	string	Country slug (FK to countries.slug)
city	string	City
founded	int	Founding year
students	int	Total enrollment (undergrad + grad)
intl_student_pct	int	International student percentage
accepts_ib	bool	Accepts International Baccalaureate
accepts_a_levels	bool	Accepts UK A-Levels
accepts_ap	bool	Accepts US Advanced Placement
website	url	Official institution website
network_strength_tier	enum	One of S/A/B/C/D
network_strength_quality	enum	full/partial/unavailable
curriculum_relevance_tier	enum	One of S/A/B/C/D
curriculum_relevance_quality	enum	full/partial/unavailable
employability_tier	enum	One of S/A/B/C/D
employability_quality	enum	full/partial/unavailable
teaching_quality_tier	enum	One of S/A/B/C/D
teaching_quality_quality	enum	full/partial/unavailable
institutional_health_tier	enum	One of S/A/B/C/D
institutional_health_quality	enum	full/partial/unavailable
student_experience_tier	enum	One of S/A/B/C/D
student_experience_quality	enum	full/partial/unavailable

countries

15 rows × 6 columns: country slug, name, flag emoji, visa info, application system, average international fee.

dimensions

6 rows × 3 columns: dimension slug, name, definition.

Quick start

from datasets import load_dataset

# Default config — universities
ds = load_dataset("brightkey/uni-rankings-2026")
print(ds["train"][0])

# Other configs
countries = load_dataset("brightkey/uni-rankings-2026", "countries")
dimensions = load_dataset("brightkey/uni-rankings-2026", "dimensions")

import pandas as pd

df = pd.read_csv("hf://datasets/brightkey/uni-rankings-2026/data/universities.csv")
print(df.head())

# Find S-tier on Employability
top_employ = df[df["employability_tier"] == "S"]
print(top_employ[["name", "country"]].head(20))

The 6 dimensions

Dimension	What it measures
Network Strength	Alumni density in target industries and post-graduate placement networks
Curriculum Relevance	How fast the institution adapts curricula to market needs
Employability	Graduate outcomes, post-study work visa pathways, employer pipelines
Teaching Quality	Student-faculty ratio, satisfaction, teaching focus
Institutional Health	Financial sustainability and revenue diversification
Student Experience	Belonging, international support, completion rates

The 5 tier levels

Tier	Meaning	Approx Distribution
S	Exceptional — global top ~5% on this dimension	~5%
A	Excellent — strong international standing	~20%
B	Strong — solid foundation	~35%
C	Good — adequate for most use cases	~25%
D	Limited data or below-average performance	~15%

Methodology

Full methodology: METHODOLOGY.md in the GitHub repo.

Editorial reviewer: Priscilla Han (BrightKey founder).

Updates: annually within 30 days of major government data releases (HESA, QILT, College Scorecard).

Citation

@misc{brightkey2026rankings,
  author       = {{BrightKey} and Han, Priscilla},
  title        = {BrightKey Independent University Rankings Dataset},
  year         = {2026},
  version      = {1.0},
  publisher    = {BrightKey},
  howpublished = {\url{https://huggingface.co/datasets/brightkey/uni-rankings-2026}},
  note         = {CC-BY 4.0; Wikidata: Q140002954}
}

License

CC BY 4.0 — share, adapt, even commercially. Just attribute.

The structured tier data is open. Rich editorial profiles (multi-paragraph university reviews, country guides) live on brightkey.co and are NOT released under CC BY 4.0 — they are copyrighted editorial work.

Contact

• Web: https://brightkey.co/contact
• GitHub Issues: https://github.com/arthurb2l/brightkey-university-dataset/issues