This is a repository of the data behind the No Ceilings site. To download all of the data in a single file, click here. Please use http://noceilings.org/data when linking to this page.
In 1995, at the UN Fourth World Conference on Women in Beijing, leaders from governments and civil society around the world came together and committed to ensuring that women and girls have the opportunity to participate fully in all aspects of life.
This year marks the 20th anniversary of that moment. The Bill & Melinda Gates Foundation and the No Ceilings initiative of the Bill, Hillary & Chelsea Clinton Foundation have joined forces to gather data and analyze the gains made for women and girls over the last two decades, as well as the gaps that remain.
This site and The Full Participation Report are the result—home to 850,000 data points, spanning more than 20 years, from over 190 countries. Through data visualizations and stories, we aim to present the gains and gaps in understandable, sharable ways—including by making the data open and easily available.
Learn more about No Ceilings here.
One of the most important aspects of this project is the compilation and collation of such a large number of statistics into a single location. We'd like to encourage students, researchers, and other curious people to do their own analyses and visualizations on topics they care about.
The data consists of ~900 topics (referred to as "indicators"), each of which covers a focus area across a set of years and a list of countries.
The list of indicators is available in several formats: HTML, Markdown, CSV, and JSON.
Each indicator has a series entry that is unique. This code is used throughout to link from the indicators metadata file to the individual JSON and CSV files described in the next section.
The data is collected from many sources. They can be seen in the source column.
The primary, secondary, and tertiary columns are used in the map to organize the indicators into something that can be reasonably browsed. Instead of presenting users with ~900 entries topics, they're first grouped by theme. Next, there's a primary name (seen on the map as the title at the top).
For instance, for the indicator named Gross enrollment ratio in secondary school, female:
- the primary category is Gross enrollment ratio
- the secondary is secondary school
- and the tertiary is female
The flavor column is either num or a semicolon-separated list of variants for this entry. In the indicators.json file, this is an anachronism from the process of how the data was integrated (from the CSV files). In future releases, this property won't be present (or will be set to null) for numeric data, and list the variants in an array (rather than requiring a split() call in your code).
For instance, for the indicator titled Can married women pursue a trade or profession in the same way as a man? (seen here), the variants are listed in the flavor column as Yes;No;Other.
This repository includes ~900 indicators, each in single files in both CSV and JSON format. Each file is named with its series code (described in the Indicators section, above).
Most data are between the years 1995-2014, but some data has 1990 entries, often for an indicator available in five year intervals (i.e. 1990, 1995, 2000...)
The columns of the CSV file are first the 3-digit ISO code (see below) for the country represented, followed by a column for each year in the data set.
Check out the enrollment and profession data mentioned in the previous section for examples.
Note that while the columns are in order from oldest to most recent, they may skip years if no data is available. Because the data is often quite sparse, columns are simply omitted.
Each file contains a single JSON object. Inside, a hash maps each 3-digit country code (see below) to another hash mapping each year to the value for that year. Years with no data are not included.
Again, the enrollment and profession indicators, this time in JSON format. To save space (these are the same files used on the production server), the JSON files aren't beautified.
To map a country from a 3-letter ISO code to its name, use this list (CSV format). The first column is the 3-digit code, the second column is the full country name. For some longer country names that fit poorly in the interface, a third column is present with the name used on the site.
The country region groupings are from the World Bank. Included in this repository is a file that maps region code to region name and another that maps region codes to a list of countries. Both are in JSON format.
We've added a few examples to help you get up and running quickly. Please modify, update, and make use of this code! Make new examples with other tools, environments, and languages. Share what you find and tell us what you've made with the data.
We (Fathom) received this data in a pair of flat files (one for numeric, one non-numeric). Where possible, we've corrected some errors, grouped the data into categories (the theme, primary, secondary, and tertiary columns), and are making this available in an accessible format to be usable for other projects.
If you find an error, file an issue and let us know. Many of the files seen here are auto-generated as part of our process of building noceilings.org. We'll do our best to update things or try to find better data sources. This is ~850,000 data points across hundreds of files: errors aren't unlikely, but you can help us keep things clear and up to date.
Last updated 14 April 2015