Home / Methodology

Methodology

This page sets the editorial line. It is the most important page on the site for assessing what we publish and what we deliberately do not.

What the dataset is

A list of institutional events. Each row is one appointment of a woman to a senior position in an institution where, at the time of the appointment, no woman had previously held the role. Each row carries the name of the appointee, the organisation, the role, the start date, the end date if applicable, and the public source documenting the appointment.

What the dataset is not

It is not a judgement of any individual’s qualifications, character, competence, achievements, or fitness for the role they hold or held. The unit of analysis is the institution, the date, and the rate at which institutional firsts have occurred. Whether a given appointee is the right person for the job is a separate question that this site does not engage with.

It is also not a comprehensive catalogue of every senior woman in every institution. It is restricted to firsts — the first woman to occupy a named senior position in a named organisation — because firsts are the events most likely to reflect the selection mechanism rather than ordinary staffing.

Scope

• Senior position. Heads of state and government, cabinet ministers, supreme- and constitutional-court justices, central-bank governors, intelligence-agency chiefs, military service heads and unified commands, presidents of major multilateral institutions, directors of major museums, presidents of major research universities, and Chief Executive Officers of major Fortune 500 firms. The full taxonomy is on the domains page.
• First. First woman in the named role at the named institution. Subsequent women in the same role are not separate rows unless otherwise noted (e.g. consecutive female directors of a regulatory agency, where the consecutive pattern is itself the analytical point).
• Selection mechanism. The dataset preserves the distinction between events reached by direct election, by parliamentary or board appointment, and by constitutional succession (vice-presidential or vice-prime-ministerial succession on death, removal, or resignation of the previous holder). Different mechanisms produce different kinds of evidence about how the office is filled, and the dataset's analyses do not collapse them.
• Geography. Coverage spans five continents. Western institutions remain the most thoroughly documented; the dataset's pre-1990 entries are predominantly non-Western (Sri Lanka 1960, India 1966, Israel 1969, Argentina 1974, Central African Republic 1975, Bolivia 1979) and the pre-2010 record is globally distributed. The 2018–2026 cluster is concentrated in Western institutions and is one of the dataset's principal observations.
• Time horizon. 1960 to present, with the principal cluster window 2018–2026. The dataset's earliest entry is Sirimavo Bandaranaike (Sri Lanka, 1960) — the world's first elected woman head of government. Pre-2010 entries are sparse and globally distributed; their inclusion lengthens the flat baseline against which the post-2018 acceleration is measured.

Sourcing

Each entry should be supported by at least one publicly verifiable source — the organisation’s own announcement, official biography, or reporting in a reputable outlet. Where multiple sources disagree on a date, the official organisational source takes precedence.

Where biographical details are included on an individual entry, they are restricted to what the appointee’s own employer, government, or biographers have made public.

Documented vs. speculative

This site distinguishes carefully between:

• Documented mechanisms — written policies, eligibility rules, and institutional commitments that demonstrably affect outcomes. Examples include institutional-investor diversity mandates from the largest pension funds and university endowments, BlackRock and Vanguard proxy-voting policies on board diversity, Goldman Sachs’ 2020 IPO diversity requirement, the Academy’s 2024 Best Picture eligibility criteria, and pipeline organisations such as the College of Europe and the World Economic Forum’s Young Global Leaders. These are verifiable from the institutions’ own publications.
• Speculative claims — assertions about coordination, intent, or hidden direction that go beyond what public sources support. The site does not publish speculative claims about named individuals. Where the broader interpretation of the data is itself speculative, it is labelled clearly as interpretation.

Pipeline-institution analysis

The dataset distinguishes between downstream events — first-woman appointments to senior political, intelligence, judicial, and corporate roles — and upstream events at the institutions that produce the candidate pool for those roles. Major research universities are the dataset's principal upstream layer: their faculties and graduate alumni populate substantial shares of every other domain on the site. The academia domain page sets out the pipeline argument in full and shows that university first-woman presidencies cluster in the 1994–2007 window — about a decade ahead of the broader institutional cluster the dataset's central thesis identifies.

The pipeline reading is one interpretation. It does not exclude other accounts of the same data, and it is open to refutation by additional pipeline institutions that show different timing.

Counterexamples

Counterexamples are tracked with the same care as cluster members. The Catholic papacy, the Bank of England governorship, the Chairman of the United States Joint Chiefs of Staff, and the Fields Medal’s record after 2014 are examples of senior positions in which the first-woman appointment has not occurred or in which the cluster trend is not observed. See counterexamples.

Editorial line

• No diagnostic or psychological claims about named individuals.
• No insinuation by association. Family backgrounds of appointees are mentioned only when documented by the appointee’s own employer or by mainstream reporting at the time of appointment, and only when directly relevant to the institutional question being analysed.
• Counterexamples receive equal editorial weight.
• The data is the artefact. Interpretation is labelled as such.

Corrections

The dataset will contain errors. Where an appointment is misdated, misattributed, or wrongly classified as a first when an earlier appointee held the role, the entry will be corrected and a brief note added to the about page. Public sources for corrections are welcome.

Update frequency

The dataset is appended as new firsts are documented and as historical firsts are added. The site is built from a single CSV; charts and indexes regenerate from that source.