Support the benchmark
The benchmark scores 125 countries on nine capabilities from public data, and its code is open. Envisioning pays for the work and asks for no money. It needs use, evidence, an outside review and people who will argue with it. The ways to help below start with the cheapest.
Anyone can read, quote and fork the benchmark for free.
Start here if you have five minutes
You know something the data does not. Each of these takes one message.
Argue with a score
Tell us where a number is wrong about the country you know.
- It needs.
- The country, the capability and what the score misses.
- Usually.
- Anyone who knows a country better than a global dataset does.
- What it changes.
- The objection is published beside the score it targets, so the disagreement travels with the number.
Name a dataset for a gap
Point us at a published series that covers a capability we cannot measure.
- It needs.
- A comparable series over at least two countries, with an open URL, a publisher, a reference period and a stated method.
- Usually.
- Statisticians, researchers, anyone who works with a national source.
- What it changes.
- A gap becomes a scored indicator, which raises confidence for every country at once.
The fastest useful thing is to close a gap. 18 indicators in the registry have no dataset behind them, and each one lowers confidence for every country at once. The open gaps list says what each is trying to measure and what a usable series would look like.
These take a piece of work
You have a case, a series or a decision to test the benchmark against. These take real work and change what the benchmark can measure.
File an evidence record
Document something a country actually delivered where the benchmark can only record a gap.
- It needs.
- One published number, the period it covers, the publisher, and a statement of what it does not show.
- Usually.
- Public servants, journalists, researchers with a case in front of them.
- What it changes.
- The record is published against the indicator it answers. It never moves a score, and it is how a capability with no dataset can still be read.
Use it against a real decision
Put the benchmark next to a decision your institution is already making, and tell us where it broke.
- It needs.
- The decision, the capability you read, and what the number failed to answer.
- Usually.
- Policy teams, development banks, schools of government, newsrooms.
- What it changes.
- A measure nobody applies stays a hypothesis. Several of the registry gaps came from exactly these conversations.
Fix the code or the data
Send a source adapter, a broken series, a wrong label or a bug.
- It needs.
- A reproducible case, or a pull request against the repository.
- Usually.
- Developers and data engineers.
- What it changes.
- A source adapter raises confidence across every country in one change.
Review the method from outside
Read how the benchmark separates capability from income, and tell us where the method fails.
- It needs.
- Familiarity with composite indicators or with regression on country data, and time to read the decision record.
- Usually.
- Statisticians, econometricians and researchers who build or audit indices.
- What it changes.
- The reading of what is left after income stays offline until someone outside the project has reviewed it. A review is what publishes it, or withdraws it.
These take an ongoing commitment
These ask an institution to stay with the benchmark for a while: to keep a country reading current, or to put the results in front of people who will argue with them.
Build a country layer
Read one country in its own language, at its own depth, beside the English benchmark.
- It needs.
- A lexicon, an institution map for that country, and somebody who can keep both current.
- Usually.
- Institutions inside the country being read.
- What it changes.
- That country gets its shape, its agenda, its institutions and its subnational spread in the language its decisions are made in.
Host a discussion of the results
Put the benchmark in front of people who will argue with it: a seminar, a working group or a newsroom.
- It needs.
- An audience that makes or studies the decisions a capability score is meant to inform.
- Usually.
- Universities, schools of government, think tanks, statistical offices.
- What it changes.
- A discussion surfaces objections and missing measures. Each objection filed is published beside the score it targets, and a missing measure can become a declared gap.
Read the limits first
Do not quote it or build on it without knowing where it fails.
- The thesis states the claim under test and shows how far the data supports it today.
- The known limits name the numbers that say more about a missing dataset than about the country.
- The objections are published beside the scores they argue with.
- Brasil, México, Colombia, Chile, Argentina, Perú, Ecuador, Bolivia, Venezuela, Paraguay, Uruguay, Costa Rica, Panamá, Guatemala, Honduras, El Salvador, Nicaragua, República Dominicana, España, Portugal, Angola, Moçambique and Guiné-Bissau have country layers, each in its own language. The structure works for any country whose institutions want to build the same thing.
The code and the data are at the repository on GitHub.
If none of that fits
Several of the gaps in the registry came out of a conversation with somebody who measures a country for a living.
Tell us what your institution measures, what it cannot measure, and which decision it would want this to inform. Write to us.