1.3 CEFR Band Descriptors & 0–100 Score Mapping

CEFR Band Descriptors & 0–100 Score Mapping

Quick Summary: The EF SET uses a continuous 0 to 100 score scale that directly mirrors the six global proficiency tiers of the Common European Framework of Reference for Languages (CEFR). Reading Comprehension and Listening Comprehension are evaluated independently on 0–100 scales and averaged equally (50/50) to produce the overall composite score. This section analyzes the exact numerical boundaries, functional receptive Can-Do descriptors, and empirical score concordance with major testing suites such as TOEFL iBT and IELTS Academic.

---

The Common European Framework of Reference for Languages (CEFR)

Established by the Council of Europe after decades of applied linguistic research, the Common European Framework of Reference for Languages (CEFR) has become the preeminent global standard for describing language proficiency. Prior to the CEFR, terms like "fluent," "intermediate," or "advanced" lacked objective definitions and varied widely between educational institutions and employers.

The CEFR organizes language competence into three broad tiers, subdivided into six distinct proficiency levels:

• Basic User: A1 (Beginner) and A2 (Elementary)
• Independent User: B1 (Intermediate) and B2 (Upper Intermediate)
• Proficient User: C1 (Advanced) and C2 (Proficient / Mastery)

Unlike traditional exams that assess language through isolated grammar drills, the CEFR defines proficiency through action-oriented "Can-Do" descriptors—describing what learners can functionally accomplish in authentic communicative situations.

---

The EF SET Continuous 0–100 Scale Architecture

The EF SET score scale spans from 0 to 100, directly mapped across the six CEFR levels. Because the underlying IRT model calculates ability on a continuous latent scale ($\theta$), the resulting 0–100 score provides a precise, granular measurement within each band.

[!NOTE] Two published presentations of the same scale. EF's CEFR overview page compresses the bottom of the scale and prints "1–30 = A1". EF's English levels conversion table and the "Understanding the results" strip on the certificate itself are more granular: 0–20 is reported below A1 — labelled A0 Novice on the certificate and pre-A1 in the conversion table — and A1 begins at 21. Use the granular version, because it is what the document an employer receives actually shows. Everything from 31 upward is identical in both presentations.

---

Subscore Computation, Composite Averaging & Asymmetrical Profiles

The EF SET generates three distinct metrics upon test submission:

1. Reading Score (0–100 scale)
2. Listening Score (0–100 scale)
3. Overall Composite Score (0–100 scale)

The Composite Formula

EF states that "the reading and listening sections of the EF SET are equally weighted in calculating the score" and that "the overall EF SET score is an average of the two section scores":

$\text{Overall EF SET Score} = \frac{\text{Reading Score} + \text{Listening Score}}{2}$

EF does not publish a rounding rule in prose, but its own sample certificate settles the question: Reading 41 and Listening 50 average to 45.5 and are reported as 45. Do not assume a half point will be rounded up in your favour, and do not assume a fractional average can lift you across a band boundary.

Analyzing Asymmetrical Proficiency Profiles

Receptive language acquisition is rarely completely symmetrical. Many candidates develop advanced reading skills through academic literature while lagging in auditory processing, or conversely, develop strong listening skills through immersion while struggling with complex written syntax.

[!IMPORTANT] Institutional Thresholds: When presenting credentials to multinational corporations or academic admissions offices, note that many institutions enforce minimum subscore thresholds alongside the composite score. For example, a global consulting firm may require an overall score of B2 (51+) with no individual subscore below 50.

---

Receptive Can-Do Functional Descriptors Across Proficiency Tiers

To bridge numerical scores with real-world workplace and academic execution, the CEFR details functional capabilities across key communicative scenarios.

---

Standardized Exam Concordance: EF SET vs. TOEFL iBT vs. IELTS Academic

EF publishes its own CEFR conversion table on the English levels page, built from each test provider's published CEFR alignment rather than from a direct equating study. EF itself calls it "a good approximation" and warns that mapping one levelling system onto another is "notoriously hard."

For TOEFL iBT, use ETS's own CEFR alignment of the 0–120 total score rather than a third-party chart: B1 ≈ 42–71, B2 ≈ 72–94, C1 ≈ 95–113, C2 ≈ 114–120.

What EF's Correlation Studies Actually Found

EF has published two peer-style correlation reports, and it is worth knowing their real scope before you quote them:

Three cautions follow directly from that table:

• Both studies used the longer PLUS form, not the 50-minute test you are preparing for. The 50-minute form has a wider standard error by design (0.45 versus 0.29).
• The correlations are in the 0.70 to 0.80 range, not the 0.85-plus figure that circulates on prep sites. EF's own report notes the two tests "may simply be getting a slightly different constellation of English language traits."
• A correlation is not a concordance. EF's report explicitly stops short of publishing a score-to-score conversion formula.

Using the Concordance Sensibly

1. The B2 Threshold (EF SET 51–60): EF aligns this band with IELTS 5.5–6.5, TOEIC 785–940 and Cambridge 160–179 — the common benchmark for undergraduate admission and multinational employment.
2. The C1 Threshold (EF SET 61–70): aligned with IELTS 7.0–8.0, TOEIC 945–990 and Cambridge 180–199 — typical of competitive graduate admissions.
3. The C2 Threshold (EF SET 71–100): aligned with IELTS 8.5–9.0 and Cambridge 200–230, above the top of the TOEIC Reading and Listening scale.

[!TIP] Leveraging Concordance for Professional Profiles: EF designs the EF SET "primarily as a tool for individuals to evaluate themselves," and it is unproctored, so it does not replace a proctored test for a student visa or an admissions requirement that names IELTS or TOEFL. Where it works well is on a CV or LinkedIn profile: state your EF SET score with its CEFR band, and cite the approximate CEFR-equivalent range rather than claiming a specific IELTS or TOEFL score you have not sat.