Data that does not exist yet
Comparable data on women’s life cycles across countries: two sides of every rule, a separate random sample in each country, and microdata in a secure environment with committee approval.
WLCE Index method and data access for researchers
Methodology, reproducible calculations and research partnership
Become a research partner How we measureWhat taking part gives a researcher
The methodology will be open to review before the first answer is collected.
The Women’s Life-Cycle Economy Index (WLCE Index) will publish its protocol, questionnaires, codebook, calculation code, the formula before the first global wave, and summary results after a disclosure-risk check. Individual participants’ answers (participant-level data, with names and contact details replaced by codes) will be available in a secure research environment under a data access agreement.
Authorship by contribution
Who did what is stated in every publication: measurement cards, the map of laws and rules, country technical reports, the analysis of the pilot, joint papers and talks at conferences.
A voice in the methodology before the data exists
Comments on the protocol, cards and formula are taken before data collection, and on the code with each release; a comment identifying a flaw gets a written answer with the revision.
An international comparative context
One methodology in all program countries gives your country and your topic a place in an international comparison, with your country’s technical report and an external statistical audit.
Work with a new standard
The Equal Consideration Standard rests on the same methodology as the Index; a research partnership can include its verification rules and its implementation pilots in organizations.
Materials for teaching and testing methods
Synthetic data and the open calculation code, published with each release of results, can be used in teaching and for testing your methods even before access to the microdata is granted.
What will be published, and when
2026: the architecture year. There are no documents to review yet: each will be recorded in the Index Results Register within seven days of publication.
The How we measure page explains the construct in detail.
The stability of positions will be checked by an external statistical auditor before every edition: the Index will be recalculated in thousands of runs in which several assumptions change at once (imputation of missing data, the aggregation formula, the weights), and for each country the median position and the range that holds 90 percent of the runs will be published. The model is the European Commission’s audit of the Global Innovation Index: in 2025, 5,000 runs and three classes of assumptions. The auditor’s report will be published together with the Index. This tests how sensitive the result is to methodological decisions. The uncertainty of a score that comes from the random selection of participants will be estimated and published separately.
| What | When |
|---|---|
| Research protocol and methodology | Before data collection |
| Measurement card for each rule (one paired measurement of the Index): the rule on paper, the question women answer about it, what the rule decides (access, allocation, compensation, protection or recognition), which women it applies to (the denominator), and the benchmark with its external source | Before data collection |
| The Index formula: what counts as 0 and as 100 points, the weights, the rules for combining measures into the overall score, and how missing data is filled in | Preliminary formula and verification plan before the pilot; version 1 after the pilot, before the first global wave |
| Criteria the pilot must pass | With the methodology note, before the pilot |
| Questionnaires in every national version | Before data collection for each wave begins, in the language of that version |
| Codebook: the list of measures and what each answer code means | With each release of results |
| Calculation code for the Index | With each release of results |
| Aggregates (summary results) after a disclosure-risk check: group sizes, rare combinations of characteristics, and the possibility of matching with other published results | With each release of results |
| Synthetic data: artificial records that copy the structure of the real ones, with invented values in place of real answers, for teaching and for testing code | With each release of results |
| Technical report for each country: how the random-sample survey (the probability sample) was built, how much precision it really gives (effective sample size and design effect), the response rate, the weights, and who was excluded and why | Before the country is compared with others |
| External statistical audit of the Index | Alongside the Index |
Why individual answers are available only in a secure
environment
In the research dataset, names and contact details will be replaced with codes.
Such data is called pseudonymized and remains personal: a combination of life circumstances can sometimes reveal whose record it is. In a study covering violence, menopause and economic dependency, the combination of country, age group, a rare life situation and employment sector can point to one person.
Researchers will analyze individual answers inside the secure environment, within an approved project and the permitted uses of the data. Downloading individual records to personal devices will be prohibited.
Before anything leaves the environment, tables, charts and other outputs will go through a disclosure-risk check. The check considers group sizes against a set minimum, rare combinations of characteristics, and the possibility of matching with other available results. It determines which materials can be released and which changes are needed to protect participants.
What happens when such data is published anyway is discussed on the Ethics and data page.
How to apply for access to participant-level data
Getting access to individual participants’ answers (pseudonymized participant-level data) will take four steps.
1 Application
You apply to the Data Access Committee: the research question, the variables you need, the analysis plan and the outputs you intend to publish. The committee will review each application and answer in writing.
2 Data access agreement
You sign it before your work begins. It rules out any attempt to trace a record to a person and any download of individual records to your own computer; linkage with outside data needs the committee’s approval.
3 Work inside the environment
The analysis runs inside the secure research environment, where the data is stored. Only outputs that have passed the final check leave it. Individual women’s records stay inside and are never copied out of it.
4 Output check
Before anything leaves the environment, each output undergoes a disclosure-risk check: group sizes against a set minimum, rare combinations of characteristics, possible matching with other published results.
The sensitive module (the questions about violence, health and money) will be kept as a separate data class: it will have its own keys, access will be granted variable by variable, and each request will get an extra review. Access to it will be narrower than to the rest of the data.
The Data Access Committee will rule on every request under rules that apply to everyone equally, without exception. Its powers will be set in writing before fieldwork begins. Individual participants’ answers will be closed to donors and to the organization’s leadership, including the founder. The committee will be able to refuse anyone, and its refusal will be final. Donors will receive only summary results.
How to review the methodology before data exists
The program will publish its protocol and measurement cards before data collection, a preliminary formula and verification plan before the pilot, and version 1 of the formula after the pilot, before the first global wave, so that each document can be criticized before there are results to defend.
Each document appears in the Index Results Register within seven days of publication. To review a draft before it is published, propose a research partnership.
Protocol and measurement cards
During methodology development, you can send comments on the design of the research and on the description of how each rule will be measured.
Pilot methodology note
Before the pilot begins, you can send comments on the plan for conducting and analyzing it, including the eight criteria by which the pilot’s results will be assessed.
National questionnaire versions
Comments on the wording, translation and clarity of questions are accepted in the language of the relevant version before data collection for the next wave of the research begins.
Calculation code
With each release of results, the corresponding version of the code will be published. Researchers will be able to check the calculations and send comments.
The calculation rules are announced before the global data comes in. The program will defend only a formula that was tested in the pilot and passed review before the global wave.
A comment that identifies a flaw will get a written answer. The answer will be published with the revision, so the record shows what was raised and what changed.
Areas of joint research work
Seven areas are open to outside co-authors.
01
Measurement cards and the indicator registry
The card for each rule and the registry that lists every indicator with its source.
02
Map of laws and rules, by country
What the law, the employer, the medical protocol or the algorithm provides in each country, rule by rule.
03
The Index’s own audit of AI systems
How the audit is designed, and how it is repeated to check that its results hold.
04
Offline collection protocol
How data is gathered in person through clinics, women’s organizations and local venues.
05
Analysis of the pilot
Whether the Index measures what it claims (construct validity), whether questions work the same way in every language (invariance), whether different coders’ results agree (coding reliability), and how stable countries’ positions are when the calculation method changes (stability of ranks).
06
Per-country technical reports
The random-sample survey in each country, its effective size, the response rate, the weights and the exclusion rules for each country.
07
Longitudinal design and the retrospective life-history module
The longitudinal design, which follows the same women across waves, and the module in which a participant describes earlier stages of her life.
Authorship follows contribution: each publication states who did what.
Limits of interpretation
The design states its own limits and fixes them now, before any data exists.
A country’s position is a range, and the weights are a choice by rule
Each score will be published with a sampling uncertainty interval, a range of possible positions from the external statistical audit, and the probability of landing in a group of countries: the share of tested calculation scenarios in which the country falls into that group. The headline figure will give every part of the Index (every subindex) the same weight: a choice by rule, not a result of calculation. An expert-weighted variant will be published alongside it, so that it is visible how much the outcome depends on that choice.
The open survey is not representative
Country estimates will rest on a separate random-sample survey in each country (the probability sample). The open survey, which women join on their own, will widen reach and help find problems to study.
One wave does not support causal conclusions
One wave does not support causal conclusions about long-term consequences. Repeated waves and the retrospective module widen what can be analyzed, and whether causal conclusions are possible depends on the design and assumptions of each specific analysis.
An AI system’s regulatory compliance does not prove the absence of bias
Compliance is a legal status. A system can meet every rule and still treat women differently.
The pilot is not the first edition of the Index
The pilot tests the questions, the collection procedures, the coding and the calculation of measures; the success criteria for each test are set in advance. The first edition of the Index will be prepared after the pilot results have been assessed.
The limitations section of every release will be written first.
FAQ
1. What will be open?
The WLCE Index methodology, questionnaires, codebook, calculation code, aggregates after a disclosure-risk check and synthetic data will be open. Individual participants’ answers will be available only inside a secure research environment under a data access agreement: pseudonymized data is still personal data.