Written and maintained by CASRAI Editorial Board
Published
On this page
Our World in Data is a free, non-profit website that publishes charts, articles and downloadable data on large global problems such as poverty, climate, health, population and conflict. Researchers use it as a starting point for secondary data analysis: most of its data comes from third-party providers such as the WHO, the UN and the World Bank, so a correct citation names both Our World in Data and the original provider.
What Our World in Data is
Our World in Data (often shortened to OWID) describes itself on its About page as a free, non-profit website. Its content is written by researchers at the University of Oxford, and the team is affiliated with the Oxford Martin Programme on Global Development. The founder named on the page is Max Roser.
It is best understood as a data publication rather than a primary data archive. It does not run surveys or field studies at global scale. It collects series from official statistical agencies, international organisations and research groups, processes them into a consistent form, and presents them as interactive charts with written context. That makes it useful for orientation and cross-country comparison, and it changes how you should cite it (see the licence section below).
Who runs it and what it covers
The site is owned, published and maintained by the non-profit Global Change Data Lab, which also maintains the data tools. According to its About page (read 2026-11-19), the site covers more than 120 topics, including poverty, CO2 and climate, migration, democracy, global health, population, armed conflicts and artificial intelligence, and hosts over 14,000 interactive charts. These counts are the publisher’s own and change as it adds material.
Content appears in three formats: long-form articles, short data insights and topic pages that collect the charts and articles on one subject. The About page also points readers to its FAQs for citation questions and to an article explaining how the team chooses which topics to work on and which metrics to provide.
How to find and download the data
Every chart on the site names its source, and the chart’s own documentation is the right place to start. The FAQ states that charts can be downloaded with the data they show, and that regional aggregations or per-capita transformations are noted on the chart or indicator page. Based on the FAQ and the developer documentation, a practical route is:
- Open the chart or topic page for your question and read the source line shown with the chart.
- Open the section headed “Learn more about this data” at the bottom of the chart, which carries the suggested citation text and the underlying provider.
- Download the data behind the chart, or use the API to request the full indicator or only the subset shown in the chart.
- Record the provider, the processing note (“processed by Our World in Data”) and the date you retrieved the file.
The same citation text is available on indicator pages, so you can cite an indicator without going through a chart.
For reproducible work, keep the downloaded file in your project archive alongside a short provenance note: the chart or indicator page address, the provider named there, the retrieval date and the licence statement shown at the time. Charts are updated as providers release new editions, so a figure fetched today may differ from one fetched next year, and a dated copy is what lets a reviewer reproduce your numbers. This is general data-management practice rather than a rule stated by Our World in Data.
Our World in Data API
The developer documentation at docs.owid.io describes four APIs (read 2026-11-19). It states that they are under active development and being consolidated, so check the documentation before building a long-running pipeline on a particular endpoint. The data table below summarises what each API is for. The FAQ adds that the chart API gives direct CSV URLs plus JSON metadata, for the full data or only the subset shown in a chart.
Licence and citation: the two-layer rule
This is the point researchers most often get wrong. The FAQ (read 2026-11-19) sets out two layers:
- Data produced by Our World in Data is published under a Creative Commons Attribution (CC BY) licence, and the About page describes charts as free to reuse under a permissive Creative Commons licence.
- Third-party data stays under the provider’s own licence terms. The FAQ tells users to check those terms before reuse.
For attribution, the FAQ says you must credit both Our World in Data and the underlying third-party data provider. Its worked example reads: “Gapminder (2015); UN Inter-agency Group for Child Mortality Estimation (2025) – processed by Our World in Data”. Use the exact text from the “Learn more about this data” section of your chart, because it reflects the specific series and year.
For wider background on licence terms for shared data, see the CASRAI guide to Creative Commons licences for research data. For the citation format of the file itself, see how to cite a dataset.
Worked example: building a citation
Suppose a chart on child mortality draws on two series. The FAQ’s own example credits the original compilers first and then Our World in Data: “Gapminder (2015); UN Inter-agency Group for Child Mortality Estimation (2025) – processed by Our World in Data”. Three things in that line matter for your reference list. The providers are named with their own publication years, the year reflects the edition the chart uses, and the closing phrase tells the reader that Our World in Data transformed the series. If you downloaded a CSV, add the page address and your retrieval date in your reference manager so the entry resolves to the exact page you used.
Checklist before you reuse a chart in a paper or slide deck
- Confirm the series is produced by Our World in Data (CC BY) or comes from a third party with its own licence.
- If it is third-party, read that provider’s terms before reuse; the FAQ says to check them.
- Credit both Our World in Data and the underlying provider, using the suggested citation text.
- Note any aggregation or per-capita transformation mentioned on the chart or indicator page.
- Keep the retrieval date and the file you used.
Strengths and limits for research use
Strengths
- Source transparency: each chart names its provider and links to the processing notes, so a reader can trace a figure back to the original series.
- Machine-readable access: CSV and JSON metadata are available through the API rather than only through screenshots.
- Breadth: one interface for more than 120 topics, which suits scoping studies and context sections of grant proposals.
- Open licensing for the data Our World in Data produces, with clear attribution wording.
Limits
- It is a secondary layer. Values may be aggregated, converted to per-capita rates or otherwise transformed. If your analysis depends on the original methodology, cite and, where possible, retrieve the provider’s own release.
- Licence terms differ by series. A chart can be reusable under CC BY while a third-party series behind it is not, so check each indicator you use.
- The APIs are described as under active development, so endpoints and formats can change.
- Topic selection is editorial. The site explains how it chooses topics and metrics, which means coverage reflects those choices rather than every possible indicator.
Is Our World in Data a reliable source?
This guide does not rate sources. What can be stated from the site’s own pages is how it makes checking possible: a named non-profit publisher, researchers at the University of Oxford writing the content, a named source on every chart, and notes where data are transformed. For a paper, treat it as you would any secondary compilation: cite the original provider, record the Our World in Data processing, and compare against the provider’s release when a number is load-bearing. The general distinction is explained in primary vs secondary sources in research, and the analysis side in secondary data analysis explained.
Alternatives and complements
The World Bank is a useful illustration of why the provider’s terms matter. Its Terms and Conditions page (read 2026-11-19) says that non-commercial use of its Materials is permitted, that derivative works and commercial uses need prior written consent from the relevant member institution, that rights are conditional on attribution to the World Bank Group and any third-party content providers, and that datasets are governed by separate terms of use incorporated by reference. In other words, the licence you inherit for a World Bank series is set by the World Bank’s dataset terms, not by the fact that it appears in an Our World in Data chart, which is the same point the Our World in Data FAQ makes about third-party data. Read the dataset terms page for the series you use.
Because Our World in Data republishes data from other organisations, the natural complement is the original provider. The About page names the WHO, the UN and the World Bank as typical sources. The CASRAI page on World Bank data covers one of them. If you are weighing where to host or discover datasets rather than consume them, see how to choose an open data repository.
When to go to the provider rather than the compiled chart is an editorial judgement, not a rule from either site: go to the provider when you need the original methodology, the full release, or a licence statement for the specific series.
Frequently asked questions
Is Our World in Data a reliable source?
The site names its publisher, cites a source on every chart and flags transformations, which lets you verify a figure yourself. Cite the original provider alongside Our World in Data and check the provider’s release for any number your argument depends on.
Who founded Our World in Data?
The About page names Max Roser as the founder. The site is owned and maintained by the non-profit Global Change Data Lab, and its content is written by researchers at the University of Oxford.
Does Our World in Data have an API?
Yes. Its developer documentation lists four APIs (Charts, Tables, Indicators and Search) and a Python library, owid-catalog, that wraps them. The documentation says they are under active development.
Can I reuse Our World in Data charts and data?
Data produced by Our World in Data is under CC BY, so reuse with attribution is allowed. Third-party data remains under the provider’s licence, and you must credit both Our World in Data and the original provider.
How do I cite Our World in Data?
Copy the citation text from “Learn more about this data” on the chart or indicator page. It names the original providers and ends with “processed by Our World in Data”, as in the FAQ example above.
Sources
- Our World in Data, About (read 2026-11-19)
- Our World in Data, FAQs (read 2026-11-19)
- Our World in Data ETL documentation, API (read 2026-11-19)
- World Bank Group, Terms and Conditions (read 2026-11-19)
Data: Our World in Data developer APIs and the data each one serves
Compilation of the four Our World in Data APIs and the owid-catalog Python library as described in the publisher's developer documentation, with the stated purpose of each, compiled 2026-11-19 from the official ETL documentation page.
Coverage: Global.
| API or tool | What it returns or does | Documentation status |
|---|---|---|
| Charts API | Primary API; returns chart data and metadata | Under active development, being consolidated |
| Tables API | Returns catalog tables produced by the ETL | Under active development, being consolidated |
| Indicators API | Semantic search of indicators | Under active development, being consolidated |
| Search API | Searches charts and datasets, like the site search | Under active development, being consolidated |
| owid-catalog (Python library) | Wraps the four APIs for use in Python | Documented alongside the APIs |
Sources
- Our World in Data ETL documentation: API (accessed 2026-11-19)
Compiled by CASRAI from the sources above. Free to reuse under CC BY 4.0 with attribution.







