Method

The evidence standard

Every fact we publish is a record, not a sentence: the figure, the source document, its publisher, its own date, the date we last checked it against the source and a confidence level. If a fact has no source we can link, it is not published.

Official and primary sources carry an entry: the State Comptroller, the Knesset Research and Information Center, ministries, regulators, national research institutes and peer-reviewed studies. Press reports can support an entry and are labelled as press on the page. Figures that come only from advocacy groups are left out until an official or primary source confirms them.

Where credible sources disagree, we show both and say which we rely on and why.

Grade A

Recent official or primary figures.

Grade B

A credible survey, or a primary source that is a few years old.

Grade C

Press or a single source. Shown only as support, and labelled.

Grade D

Asserted, not measured. Never published as a figure.

Confidence: high

Checked against the primary document and current.

Confidence: medium

Sound, but dated or resting on one source. Shown with that caveat.

Confidence: low

Never presented as a finding or an instruction.

The recency rule

What we highlight is chosen by how close it is to arriving now. Evidence about arrivals from 2023 to 2026 is featured and 2020 to 2022 is supporting; anything earlier appears only as background data or to show that a problem has persisted.

Every page shows when it was last reviewed and when the next review is due. A figure past its review date is marked as such rather than presented as current.

Starting tiers

Two families with the same plan can have very different aliyot, mostly because of what they arrived with. So outcomes are always read against one of three starting tiers, built from two tests.

Secured income: an Israeli job offer, a kept remote job, a pension or a running business, converted to shekels and at least 50% of monthly costs.

Cushion: savings plus money still to transfer that covers 12 months of costs or more.

Tier A

Both secured income and a cushion.

Tier B

One of the two: secured income or a cushion, not both.

Tier C

Neither secured income nor a cushion.

Not stated

Not stated: we do not know enough about the starting point to place it.

Consent tiers

When families share their experience, they choose how it is used and can move to a more private level at any time before publication. Consent is asked at intake, confirmed after the conversation and confirmed again with the final text in hand. Nobody is paid more for a more public story.

Data only (the default)

Nothing individual is published. The answers count only towards totals that meet the small-group rule.

Composite

A pattern drawn from three or more similar families, clearly labelled as a composite.

Pseudonym

One story under an invented name, with minor details changed and the changes disclosed.

Named, text only

A real name and quotes, with the final text approved by the person.

Named on camera

Face or voice, only after a separate second consent.

De-identification

No count, average or percentage is published for a group smaller than 11. Nor is any table that would let a reader work one out from the others. In the open-data tables this means any figure under 11 is shown as "fewer than 11" or left out, and where a hidden figure could still be worked out by subtraction (from a row total, a column total or a second table), further figures are withheld until it cannot.

Direct identifiers (names, ID and phone numbers, addresses, employers, schools, licence numbers) are removed. Details that could identify someone in combination are coarsened: a small town becomes its region, an exact date becomes a year and a rare profession becomes its profession family.

Before anything is published, a second reader asks whether a neighbour, an employer or a caseworker could recognise the family. If so, it is coarsened further or becomes a composite. Children in experiences shared with us are never named, pictured or described by age, school or diagnosis.

Right of reply

Every institution named in a Barriers Register entry is sent the entry when it is published and can reply or ask for a correction at any time. Replies are published in full with the entry. From version 1, a new finding goes to the institution 10 working days before it is published.

Institutions are named as the owners of a fix, not as villains. Private individuals are never named without their written consent.

Corrections

When we get something wrong, we correct it and log the correction below with the date, what it said, what it says now and why. The site-wide corrections log covers the rest of Aliyahu.

Data sources

These are the sources the published research draws on today. Each page links the exact document behind each figure.

• New immigrants by year (record-level open data, 2015 to 2025)

  Ministry of Aliyah and Integration, data.gov.il

  Where olim land: arrivals by year, first locality, country of origin and age.

  Other (Open). Attribution: Ministry of Aliyah and Integration, data.gov.il.

• Statistical Abstract of Israel 2025, immigration tables

  Central Bureau of Statistics

  The official national count, shown next to the Ministry records for comparison.

  Cited as published. CBS reuse terms not yet reviewed.

• State Comptroller audit reports

  State Comptroller of Israel

  Barriers Register: audited figures and findings.

  Cited as published.

• Research papers for Knesset committees

  Knesset Research and Information Center

  Barriers Register: official figures prepared for Knesset members.

  Cited as published.

• Applied social research reports

  Myers-JDC-Brookdale Institute

  Barriers Register: survey findings on recent olim.

  Cited as published.