Nuristan Education Initiative Donate
Figures & sources

Every number on this site, and where it came from.

A figure without a source is an assertion. This page lists every quantity published on nuristaneducation.org together with its origin, the method used to derive it, the date its source speaks to, and — where they exist — its limits. Every one of the entries below carries all four. Anything we cannot source this way does not go on the site.

22

Published figures

Every quantity on the site, each with its origin, method, date and limits.

8

Independent sources

Counted by organization. Each is named, dated and linked in the entry that uses it.

4

Figures withdrawn

Published, then found unsupportable, and recorded here rather than quietly deleted.

The standard

Four kinds of figure, four kinds of proof.

Last reviewed: August 2026

Derived

Computed by us from an open dataset. We publish the dataset, the method, and the script, so anyone can reproduce the number.

Cited

Taken from a third party. We name the organization, link the document, and record the date it was published — not the date we read it.

Internal

From the Initiative's own records, under the Program Data Collection Plan. Structural facts only until the first reporting cycle closes.

On the public record

Checkable against an independent registry or a governing document, without taking our word for any part of it.

The index

Every figure, by kind.

Derived

Geography and infrastructure.

Computed from open datasets, and reproducible. The script that produces every figure in this group is committed in the repository as tools/geo_facts.py.

9,579 km² — the province, by area

Source
OCHA Common Operational Datasets, Afghanistan administrative boundaries (COD-AB v03), admin level 1, pcode AF16, boundary valid on 1 June 2025.
Method
sinusoidal equal-area projection about the polygon centroid.
Corroboration
reproduces OCHA's own area_sqkm attribute (9,579.4948) to four decimal places.
As of
1 June 2025 — the boundary’s own validity date in the COD-AB release.

8 — districts

Source
the same COD-AB release, admin level 2, features whose parent is Nuristan.
Method
a count. Barg-e-Matal, Du Ab, Kamdesh, Mandol, Nurgaram, Parun, Wama and Waygal.
As of
1 June 2025 — the same release.

901 km — road and track mapped across it

Source
OpenStreetMap road data for Afghanistan (HOTOSM export), clipped to the province — 621 segments.
Method
geodesic length, haversine on a sphere of radius 6,371.0088 km, summed over every segment.
Limit
OpenStreetMap coverage in remote Afghanistan is incomplete. This is a floor, not a census.
As of
6 August 2026 — the OpenStreetMap snapshot the HOTOSM extract was cut from.

6.9 km — recorded with a paved surface

Source
the same extract, surface tags — asphalt 5.8 km plus concrete 1.1 km.
Method
summed by tag, then expressed against the mapped network as a share — 6.9 ÷ 901 = 0.77%, the form the figure takes on the home page.
Limit
201 km, 22% of the mapped network, carries no surface tag at all. The honest claim is what the data records, not what exists on the ground; we do not say that only 6.9 km of the province is paved.
As of
6 August 2026 — the same snapshot.

0 — roads classified above “secondary”

Source
the same extract, highway tags. No segment is classified motorway, trunk or primary. The highest class present is secondary, at 114 km; tertiary accounts for 455 km, half the network.
As of
6 August 2026 — the same snapshot.

94 km — of road and track per 1,000 km²

Source
the two figures above, divided.
Method
901 km ÷ 9,579 km², expressed per thousand — 94.06, published rounded to two significant figures.
Limit
it inherits the limit of the road figure exactly. Because the 901 km is a floor rather than a census, the density is a floor too. It is published because it gives 901 km a scale a reader can hold, not because it measures anything the road figure does not.
As of
6 August 2026 for the road length, 1 June 2025 for the area. This figure combines two vintages fourteen months apart. The boundary is stable and the road network is not, so the density moves with the roads.

Figures we hold but do not publish

The same computation shows 597 of 621 segments unnamed. It is true and it is not on the site — it adds nothing a reader cannot already see in the figures above.

Cited

Higher education in Afghanistan.

Third-party figures. We name the body, date the publication, and say where corroboration exists.

166 — accredited higher education institutions

Source
a third-party compilation of the Ministry of Higher Education’s accredited list — 39 public and 127 private. We could not obtain the list from the Ministry directly.
Corroboration
UNESCO recorded 128 private institutions registered with the de facto Ministry as of April 2023, which agrees with the private count.
Limit
the public count rests on a single compilation of the Ministry list and should be treated as indicative rather than official.
As of
15 August 2026 — when the compilation was retrieved. It carries no date of its own, which is part of why the count is flagged as indicative rather than official.

1 — of them in Nuristan

Source
the same accredited list, where the province’s single entry appears at item 39 among governmental institutions, named there as the Nuristan Institute of Higher Education.
Note
the institution’s own site now presents it as Nuristan University. We have found no Ministry record of a reclassification, so we report the change in how it describes itself rather than assert a change of tier. What it publishes is two faculties, Sharia and Education, with further programs described as a plan; no postgraduate program is listed. The claims elsewhere on this site rest on that published provision, not on the tier label.
Limit
we make no claim about how many private institutions operate in the province. We could not obtain a complete provincial register, and absence from an incomplete list is not evidence of absence.
As of
15 August 2026 for the accredited list; the institution’s own site checked 27 August 2026.

The prohibition on girls' education

Source
UNESCO, 11 August 2026 — “Afghanistan is the only country in the world to formally prohibit access to education for girls and women beyond the primary level,” with an estimated 2.4 million girls excluded from secondary education.
Limit
UNESCO does not publish the method behind the 2.4 million estimate, so we attribute it rather than assert it.
As of
11 August 2026 — the date of the UNESCO statement quoted.

1895–1896 — annexation, and the name Kafiristan

Source
the standard historical record of the region's conquest under Amir Abdur Rahman Khan, its incorporation into the Afghan state, and its renaming from Kafiristan to Nuristan; summarised in Minority Rights Group's profile of the Nuristanis. Dates and names, not measurements.
As of
15 August 2026 — when the Minority Rights Group profile was retrieved. The events themselves are fixed; this records when we last checked the source.

Kati, Kalasha-ala, Ashkun and Prasun

Source
Minority Rights Group, profile of the Nuristanis, which names these varieties and records that they form a branch of Indo-European distinct from the languages of neighbouring communities; language identities cross-checked against Glottolog and ISO 639-3.
Note
On the names. Kalasha-ala — the speakers' own name for the language, Kalaṣa-alâ — is the same language often catalogued as Waigali, ISO 639-3 wbk, with roughly 10,000 speakers. Strictly, Waigali names the dialect of the upper Waigal Valley rather than the language as a whole, so we use the endonym. Not to be confused with. Kalasha-mun (kls), an Indo-Aryan language of Chitral in Pakistan. The two are unrelated despite the shared name.
Limit
we deliberately do not state a count. Classifications of the Nuristani branch differ — Tregami and Zemiaki are also generally treated as separate languages within it — so naming four of them without claiming they are all of them is the defensible form.
As of
15 August 2026 — the same retrieval.

“Long-term neglect and discrimination”

Source
Minority Rights Group, quoting a provincial governor's description of Nuristan's treatment by the state.
Note
attributed as a characterisation by an official, not asserted by us as a finding.
As of
15 August 2026 — the same retrieval.
Internal

The Initiative's own records.

Structural facts about how the programs are built and run. No outcome figures appear here yet, and that is deliberate — see how we will know it is working.

Operating since spring 2023

Basis
the Initiative’s own program record. The village after-school program opened in spring 2023 under private financing and has run continuously since.
As of
21 August 2026 — the most recent board record. The spring 2023 opening is fixed and needs no currency; what this row dates is the claim that operation has been continuous since.

6–9 months — the Kabul preparation cycle

Basis
program design. The cycle length varies with the Kankor calendar.
As of
August 2026 — Program Data Collection Plan v1.2, board draft.

Roughly ages 11 to 18

Basis
the middle- and high-school grades the village program serves. A range, deliberately imprecise.
As of
August 2026 — Program Data Collection Plan v1.2, board draft.

Monthly attendance and progress checks

Basis
the Program Data Collection Plan, which sets a fixed monthly reporting cycle against a written specification for each program.
As of
August 2026 — Program Data Collection Plan v1.2, board draft.

September 2026 and September 2027

Basis
a commitment, not a measurement. Monthly data collection begins in September 2026; the first complete cycle of results reports twelve months later. Hold us to it.
As of
21 August 2026 — the date the board made this reporting commitment.

At least five university admissions a year

Basis
a stated target approved by the board. It describes what we are aiming at, not what has happened.
As of
mid-June 2026 — the date the board approved the target.
Background reading

Where the region has been studied properly.

Listed separately from the figures above, and deliberately so: no number on this site is drawn from these works. They are here because a reader who wants to understand Nuristan rather than audit us should not have to start from a search engine.

Schuyler Jones, Men of Influence in Nuristan (1974)

A Study of Social Control and Dispute Settlement in Waigal Valley, Afghanistan. London and New York: Seminar Press, 1974. Seminar Studies in Anthropology 3; xiv, 299 pp. ISBN 978-0-12-785393-2. Reviewed in The Journal of Asian Studies and American Anthropologist (1976).

Jones lived and worked in the Waigal Valley and wrote the standing ethnographic account of how authority, obligation and dispute settlement actually functioned there — how the valley governed itself, in a province governed from a distant centre. It is also the valley whose language, Kalasha-ala, appears in the entry above. Anyone assessing this Initiative who wants to know what kind of place it works in should read Jones before reading us.

Corrections

Figures we have removed.

This site has carried numbers it could not support. Rather than quietly delete them we record them here, because an organization's willingness to publish its own corrections is worth more than a page of figures that has never changed.

If you find a figure on this site you believe is wrong, write to info@nuristaneducation.org and we will check it, correct it if you are right, and record the change here.

Removed, August 2026
  • “Roughly 85% of expenditure directed to teaching and student support” — traced to no ledger line
  • “More than eight hours by road” — no citable source, and two pages disagreed on the destination
  • A claim that program students were outperforming comparable non-participants — self-reported, no methodology, and no measurement system yet exists to support it
  • An undated UNESCO figure, since replaced by the dated August 2026 estimate

Every figure, checkable.

That is the standard this page exists to hold us to. If we cannot source it, we do not print it.

Donate securely Accountability