The score scale at a glance
The updated TOEFL iBT reports four section bands and one overall band, all on a 1–6 scale in half bands:
- Reading: 1–6
- Listening: 1–6
- Writing: 1–6
- Speaking: 1–6
- Overall: average of the four, rounded to the nearest half band
The test has four sections in a fixed order, with no scheduled break:
| Section | Format | Scoring basis |
|---|---|---|
| reading | about 30 min, 50 items | selected-response items, multistage adaptive |
| listening | about 29 min, 47 items | selected-response items, multistage adaptive |
| writing | 23 min, 12 items | 10 objective items + 2 scored responses |
| speaking | about 8 min, 11 items | 11 scored spoken responses |
How Reading and Listening are scored
Adaptive scoring: routing module first, then second module
Both reading and listening use multistage adaptive testing.
- You start with a routing module.
- Your performance there selects an easier or harder second module.
- ETS then converts your performance to a section band.
The practical consequence is simple:
- The harder second module is what gives access to 5.5–6.
- If you miss too much in the routing module, you can still score well, but the top bands usually become unreachable.
What counts and what does not
- Every scored selected-response item is worth 1 point.
- There is no penalty for guessing.
- Some items are unscored pretest items and are not identified.
What matters most for top bands
For learners targeting 5.5–6, the routing module is the highest-leverage part of the test.
A good rule:
- Band 6 target: near-perfect routing performance, then very high accuracy in the harder second module
- Band 5.5 target: strong routing performance, then solid control in the harder second module
- Band 5 target: good overall accuracy, even with some losses in routing or second module
How Writing and Speaking are scored
Unlike Reading and Listening, writing and speaking are linear: everyone on the same form gets the same tasks.
Writing: raw score out of 20
| Task | Count | Marks per task | Raw total |
|---|---|---|---|
| Build a Sentence | 10 | 1 | 10 |
| Write an Email | 1 | 0–5 | 5 |
| Write for an Academic Discussion | 1 | 0–5 | 5 |
| Total | 12 items | 20 |
A few score implications:
- Build a Sentence supplies half of the raw Writing marks.
- The two extended responses supply the other half.
- A very strong objective score can protect you from a slightly weaker essay score.
Example:
- Build a Sentence: 10/10
- Write an Email: 5/5
- Write for an Academic Discussion: 4/5
- Total = 19/20, which is a Courselo estimate of Band 6
Speaking: raw score out of 55
| Task | Count | Marks per response | Raw total |
|---|---|---|---|
| Listen and Repeat | 7 | 0–5 | 35 |
| Take an Interview | 4 | 0–5 | 20 |
| Total | 11 items | 55 |
This is the most important weighting fact in the whole test:
- Listen and Repeat = 35/55 raw points = 64% of Speaking
So if you want the fastest band gain in Speaking, train exact repetition accuracy, rhythm, and clean endings first.
Who scores the productive tasks?
ETS states that the written and spoken responses are scored with AI scoring, and ETS materials also mention certified human raters. The exact split is not publicly verified, so do not base strategy on assumptions about “AI-only” or “human-only” scoring.
How the overall band is calculated and rounded
The overall score is the mean of the four section bands:
Where:
- = Reading band
- = Listening band
- = Writing band
- = Speaking band
Rounding rule
ETS rounds to the nearest half band.
Examples:
- 5.125 → 5
- 5.25 → 5.5
- 5.75 → 6
What you need for each overall target
| Overall target | Minimum average needed | Typical section profiles |
|---|---|---|
| 6 | 5.75 | 6 / 6 / 5.5 / 5.5 or 6 / 6 / 6 / 5 |
| 5.5 | 5.25 | 5.5 / 5.5 / 5 / 5 or 6 / 5.5 / 5 / 4.5 |
| 5 | 4.75 | 5 / 5 / 5 / 4 or 5.5 / 5 / 4.5 / 4 |
| 4.5 | 4.25 | 4.5 / 4.5 / 4 / 4 |
| 4 | 3.75 | 4 / 4 / 4 / 3 |
What the bands mean: CEFR, old 0–120 comparison, IELTS
Until January 2028, score reports also show a comparable score on the old 0–120 TOEFL scale.
| TOEFL iBT band | CEFR | Comparable old TOEFL total | Approx. IELTS |
|---|---|---|---|
| 6 | C2 | 114–120 | 8–9 |
| 5.5 | C1 | 107–113 | 7.5 |
| 5 | C1 | 95–106 | 7 |
| 4.5 | B2 | 86–94 | 6.5 |
| 4 | B2 | 72–85 | 6 |
| 3.5 | B1 | 58–71 | — |
| 3 | B1 | 44–57 | — |
| 2–2.5 | A2 | lower | — |
| 1–1.5 | A1 | lower | — |
Practical meaning of the bands
| Band | Practical profile |
|---|---|
| 6 | Near-complete control. Errors are rare and usually minor. You can handle top-difficulty material across all four sections. |
| 5.5 | Strong C1. Clear, accurate, and reliable performance with only occasional lapses under speed or density. |
| 5 | Solid C1. Good academic readiness, but with noticeable losses in precision, range, or consistency. |
| 4.5 | Upper B2. Generally effective, though weaker control shows in dense listening/reading or timed production. |
| 4 | B2 minimum. Functional for many university contexts, but with clear limits in speed, range, and accuracy. |
MyBest scores, validity, and institution requirements
MyBest
ETS reports MyBest scores, which combine your best section scores from all valid TOEFL iBT tests taken in the last 2 years.
That means a university may see:
- your test-date score profile, and/or
- a stronger combined profile built from multiple sittings
Other score-report facts
- Scores are typically available in about 3 days.
- Scores remain valid for 2 years.
- You may retake after the 3-day rule.
- There is no universal pass mark; institutions set their own overall and section minimums.
How Courselo estimates section bands on mocks
Courselo mocks are fixed-form: they simulate the routing module + harder second module. Because every item counts in the mock, the conversion is only an estimate of real ETS scaling.
Reading and Listening estimates
| Section band | Reading estimate | Listening estimate |
|---|---|---|
| 6 | 93% | 92% |
| 5.5 | 86% | 85% |
| 5 | 77% | 75% |
| 4.5 | 68% | 66% |
| 4 | 57% | 56% |
| 3.5 | 45% | 44% |
| 3 | 33% | 32% |
Writing and Speaking estimates
| Section band | Writing raw / 20 | Speaking raw / 55 |
|---|---|---|
| 6 | 19 | 53 |
| 5.5 | 17 | 48 |
| 5 | 16 | 43 |
| 4.5 | 14 | 38 |
| 4 | 12 | 35 |
| 3.5 | 10 | 29 |
| 3 | 8 | 24 |
Where the marginal points are cheapest to gain
Highest-return areas
- Complete the Words: about 60% of Reading items
- Listen and Repeat: 64% of Speaking raw points
- Build a Sentence: 50% of Writing raw points
Best scoring strategy by target
If you are aiming for 4.5–5
- Get Build a Sentence close to perfect.
- Push Listen and Repeat to consistent 4s.
- Protect the routing modules in Reading and Listening.
- In writing, cover all required points before trying to sound advanced.
If you are aiming for 5.5–6
- Treat the routing module in both adaptive sections as decisive.
- Train zero-error routines in Complete the Words and Build a Sentence.
- In Speaking, raise Repeat responses from 4 to 5; one-point gains there compound quickly.
- In the two writing tasks, eliminate content omissions and grammar breakdowns first.
Bottom line
Scoring on the updated TOEFL iBT is simple in structure but not in scaling:
- Reading and Listening: 1 point per scored item, adaptive scaling, harder second module needed for top bands
- Writing: raw 0–20
- Speaking: raw 0–55
- Overall: average of the four section bands, rounded to the nearest 0.5
For most learners, the fastest score gains come from the high-weight, high-control tasks: Complete the Words, Build a Sentence, and Listen and Repeat. If those are strong, your section bands rise more cheaply than by trying to improvise your way to better scores in the lower-weight tasks.