What Is STANAG 6001?

STANAG 6001 is the NATO standard for measuring language proficiency. It uses a scale of 0 to 5 across four skills — listening, speaking, reading, and writing. The standard is used across NATO member nations, EU institutions, and organisations such as Frontex, Interpol, Europol, and NATO Centres of Excellence.

Tests based on this standard are high-stakes. Results are used to make decisions about employment, deployment, promotion, course admission, and proficiency pay. Getting the measurement right matters enormously — for individuals and for operational safety.

What Is the Purpose of a STANAG 6001 Test?

Every STANAG 6001 test must accomplish three things:

1. Assess a person's unrehearsed, spontaneous language ability in real-world communication situations — not memorised responses or textbook language.
2. Determine which proficiency level best describes what that person can do consistently and sustainably.
3. Report that level accurately to the relevant stakeholders.

What Makes STANAG 6001 Tests Different?

STANAG 6001 tests are criterion-referenced, not norm-referenced. This means your result is measured against a fixed standard — not compared to how other candidates performed. Either you meet the level or you do not.

Scoring is also non-compensatory. A strong vocabulary does not cancel out weak accuracy. Doing well on Level 3 items does not compensate for failing Level 2 items. Every element of the standard must be met independently.

How Tests Must Be Designed

BILC sets out clear requirements for any test that claims to measure STANAG 6001 proficiency:

• Each of the four skills — listening, speaking, reading, and writing — must be tested and scored separately.
• Each proficiency level is treated as a separate construct and tested independently.
• Tests must include a representative sample of tasks at each level — covering a variety of topics, text types, and communication situations.
• For speaking and writing, tests must establish both a floor (what the candidate cannot yet do) and a ceiling (where their ability runs out). A result that does not clearly show both is considered non-ratable.
• Reading texts for Level 1 should average around 50 words; Level 2 around 150 words; Level 3 around 300 words.
• Listening texts should not exceed 45 seconds at Level 1, 60 seconds at Level 2, and 90 seconds at Level 3 — to avoid testing memory rather than comprehension.

What Counts as a Valid Test Text?

For Levels 2 and 3, test materials must use authentic texts — real written or spoken language produced for genuine real-world purposes, not created specifically for language teaching or testing. Level 1 may use semi-authentic texts, provided they are accepted as natural by educated native speakers.

Texts must represent the variety of English actually used in NATO contexts, not a single accent or register.

How Are Tests Written and Reviewed?

Item writers — the people who write the exam questions and speaking prompts — must work as a team, not individually. Before writing begins, the team must agree on what is being tested and how.

Every item goes through a formal moderation process before it can be used. A review panel of up to six people — including external reviewers — checks that each item aligns with the correct level, is free of errors, and genuinely measures what it is supposed to measure. Items are either approved, revised, or discarded.

Test developers are required to write 40 to 60 percent more items than needed to allow for items that do not pass moderation or that become outdated.

How Are Speaking and Writing Tests Rated?

Tests of productive skills — speaking and writing — must be rated by at least two trained and normed raters. Speaking tests must be recorded so that a third rater can resolve disagreements.

Raters go through a process called rater norming: before a major testing session, they practise rating real samples together, discuss their decisions, and align themselves to the STANAG 6001 standard. This ensures that a candidate receives the same score regardless of which rater assesses them.

Signs that a rater may be drifting from the standard

• Severity: Awarding scores that are consistently too low across all candidates.
• Leniency: Awarding scores that are consistently too high.
• Halo effect: Allowing outside impressions — such as a candidate's classroom reputation — to inflate scores.
• Central tendency: Clustering all scores around the middle, regardless of actual performance.

Testing New Exam Materials Before They Go Live

Before any new test reaches candidates, it goes through two stages of trialling:

1. Piloting — a small group, which may include native speakers and language professionals, reviews the materials in an informal setting and provides qualitative feedback on clarity, level appropriateness, and task effectiveness.
2. Pretesting — the test is administered to a larger, representative group under real testing conditions. The resulting data is analysed to decide whether each item should be kept, revised, or removed.

How Tests Must Be Administered

BILC requires that all candidates sit the test under identical, standardised conditions. Key requirements include:

• Anonymity of candidates — codes are used instead of names.
• Teachers may not test their own students.
• Speaking testers must not conduct more than eight speaking tests per day.
• Test materials must be kept secure before, during, and after the session.
• Clear written policies must exist for cheating, appeals, and disruptions.
• Candidates must be given familiarisation guides in advance, including sample questions and format information.

Summary: What the Standard Demands

In plain terms, BILC is setting a high bar for everyone involved in STANAG 6001 testing — and for good reason. The results of these tests affect people's careers, their deployments, and in some cases operational safety. The document makes clear that:

• Only purpose-built STANAG 6001 tests can produce valid STANAG 6001 scores.
• Every skill is tested separately and scored independently.
• Tests must reflect real-world language use, not academic or rehearsed performance.
• Rating must be consistent, trained, and free from individual bias.
• Candidates deserve transparent, standardised, and fair test conditions.