An interactive rebuild of Pearson's official scoring document — written so students, teachers and institutions can understand weighting, integrated skills, the Skills Profile, rescoring and every common misconception in one place.
This page rebuilds Pearson's official PTE Academic — Scoring Information for Teachers and Partners as an interactive guide. It exists to give students, teachers and institutions an accurate, jargon-free explanation of how PTE Academic is actually scored — and to bust the most common misconceptions before they cost you a band.
Whether you are about to sit the test, coach students for it, or interpret a score report — by the end of this page you will understand question weighting, integrated skills, the Skills Profile, rescoring and the truth behind the Overall score.
Overall score 10–90
Reported on the Global Scale of English
4 communicative skills
Listening · Reading · Speaking · Writing
22 question types
Each weighted by its real-world importance
Skills Profile
A diagnostic — NOT a score report
“Every question is scored differently — and we'll show you exactly how.”
01Question scores
Each of the 65–75 questions earns either correct/incorrect points or partial credit, depending on the task type.
02Question difficulty
Every question is pre-tested. Harder questions count more — that way, two test takers with different question sets are still compared fairly.
03Question weighting
Each question type contributes a different share to each score, so the test consistently rewards the skills that matter in real-world communication.
Important: Overall, Listening, Reading, Speaking and Writing are each calculated independently. The Overall score is based on every question in the test; each skill score is based only on the questions that assess that skill — and some questions count toward more than one (integrated skills).
Extended Speaking & Writing answers are scored on several traits. Two of them — Content (how appropriate your response is to the prompt) and Form (formal characteristics like word count) — act as gatekeepers. Score zero on either and the rest of the response earns no points either.
Source: Pearson — "PTE Academic Scoring Information for Teachers and Partners". Percentages are averages on a typical test.
“The weighting table is indicative — your performance is what truly drives the score.”
Under this belief, every communicative skill (Speaking, Writing, Reading, Listening) carries exactly 25% of your Overall score, and the question types inside each skill split that 25% equally between themselves. Read Aloud and Describe Image both worth ~5%. Write from Dictation and Highlight Correct Summary both worth ~3%. A simple, clean — but incorrect — interpretation.
Imagined contribution per question type (each skill capped at 25%)
Figure 1. Incorrect interpretation: Overall score as a simple average of the four skills (each skill 25%). This is not an accurate representation of how PTE calculates your score.
Question types aren't equally weighted
Describe Image alone is ~15% of Overall and ~31% of Speaking. Multiple Choice (Single Answer) is under 1%. They never count the same.
Skills aren't 25% each of the Overall
The Overall is calculated holistically across every question — not as a 25/25/25/25 average of skill scores.
Integrated tasks count toward multiple skills
Summarize Written Text contributes to both Reading and Writing. Repeat Sentence contributes to Listening and Speaking. They can't sit in a single 25% bucket.
Question difficulty matters
Two test takers seeing different question sets are compared fairly because every item is calibrated. Equal % per task ignores this entirely.
The Overall score is built question-by-question. Each task has its own pre-calibrated weighting (its real share of the Overall) and feeds only the skills it tests. Some tasks — the integrated ones — count toward two skills at once, with different weights in each. Below is the full reality, sorted by how much each task actually contributes to your Overall score.
Each column = a skill (or Overall). Each segment = a question type's share of it. Hover for details.
Figure 2. Correct interpretation: every task carries its own weighting, and integrated tasks (multi-coloured bars) feed two scored skills at once.
Weightings reflect real-world value
Describe Image (15%) and Summarize Group Discussion (9%) carry the biggest share because they exercise extended speaking with full discourse.
Integrated tasks count twice
Summarize Written Text feeds Reading (23%) AND Writing (28%). Repeat Sentence feeds Listening (17%) AND Speaking (16%). One performance, two scored skills.
Question difficulty is baked in
Each item is calibrated. Two students with different question sets get a fair Overall because difficulty is part of the algorithm.
Overall ≠ average of skills
All four skill totals can be below 90 while the Overall is 90 — because the Overall is calculated holistically across every question, not averaged.
Some PTE questions deliberately test more than one skill at once. Summarize Written Text feeds both your Reading and Writing. Summarize Group Discussion feeds both Listening and Speaking — and both also count toward your Overall score.
That's why averaging the four skills would double-count integrated tasks. The Overall is calculated directly from every question — each question counted once.
Appears in Speaking & Writing · Scores Listening + Speaking
Appears in Speaking & Writing · Scores Listening + Speaking
Appears in Speaking & Writing · Scores Listening — lives in the Speaking & Writing section but doesn't add to that skill
Appears in Speaking & Writing · Scores Listening + Speaking
Appears in Speaking & Writing · Scores Reading + Writing
Appears in Listening · Scores Listening + Writing
Appears in Listening · Scores Listening + Reading
Appears in Listening · Scores Listening + Reading
Appears in Listening · Scores Listening + Writing
“Skills Profile shows strengths — not your scaled 10–90 score.”
Replica of your official myPTE score report — click any row to see the contributing questions
Scores Overview
Your Skills Profile
Question types — grouped by the skill they score
Contributes ≈ 54% of overall scoreSpeaking
Writing
Reading
Listening
Question types — grouped by the skill they score
Contributes ≈ 16% of overall scoreSpeaking
Writing
Listening
Question types — grouped by the skill they score
Contributes ≈ 18% of overall scoreWriting
Reading
Listening
Question types — grouped by the skill they score
Contributes ≈ 5% of overall scoreWriting
Listening
Question types — grouped by the skill they score
Contributes ≈ 36% of overall scoreSpeaking
Listening
Question types — grouped by the skill they score
Contributes ≈ 11% of overall scoreSpeaking
Listening
Question types — grouped by the skill they score
Contributes ≈ 42.5% of overall scoreSpeaking
Writing
Reading
Listening
Question types — grouped by the skill they score
Contributes ≈ 31% of overall scoreSpeaking
Reading
Listening
Visual replica of the official Pearson myPTE Skills Profile panel — bar lengths show relative strength, not a score out of 90.
“Rescoring is a re-evaluation by trained human raters — not a sign of error.”
Technical review
Checks for technical issues with the audio recordings. If issues are found, the test fee and review fee are refunded. Scores do not change.
Human rescore
Every open-ended response is re-evaluated entirely by two trained human raters per response (third rater adjudicates disagreements). The new score replaces the original — it can go up, down or stay the same. One rescore per test attempt.
Appeal
Verifies that the rescore was carried out per Pearson's procedures. It does NOT re-assess responses and does NOT provide individual feedback.
Why scores can change
Rescoring is fully human. Small differences in individual rater judgments add up across the test — most rescores closely match the original, but noticeable changes happen in rare cases. A change does not mean the original was wrong; it reflects the inherent subtleties of full human evaluation.
Transparency & security
To protect the test, Pearson does not publish answers, algorithms or per-response rationales. The scoring model is proprietary — no third party has access to question difficulty data or scoring algorithms. Only official PTE materials reflect the real scoring model.
“9 beliefs we hear weekly — and the official facts that replace them.”
"I think my Overall score is inaccurately calculated; because when I calculate it, it makes a different average."
Example: Overall score of 81 (Listening 69, Reading 63, Speaking 90, Writing 64).
The Overall score is not the average of the individual skill scores. Each question type contributes directly to the Overall score, with its own different weighting. Calculating an average of the four skills will not reflect the true Overall score.
"If my Overall score is 90, then all my individual communicative skills scores must also be 90."
Example: Overall 90 (Listening 90, Reading 69, Speaking 90, Writing 90).
The Overall score is not an average of the four skills — it is calculated holistically across every question. "90" does not mean a perfect score; it means the test taker performed at the top of the range PTE measures. With exceptional performance on the heavily-weighted question types, an Overall 90 is possible even when one skill is below 90.
"The Skills Profile bars appear full, but my skill scores are not all 90. (Or: all my skills are 90, but the Skills Profile bars don't appear full.)"
Example:
Scores Overview
Your Skills Profile
The Skills Profile shows your relative strength across language categories — it is NOT scaled on the 10–90 Global Scale of English (GSE) used for skill scores. A full bar does not equal 90, and a score of 90 does not guarantee a full bar. Use the Skills Profile as a diagnostic tool, not a score report.
"Weighting percentages allow you to predict exact scores."
No. PTE Academic scores reflect your language proficiency based on three things together: your performance, question difficulty and question weighting. The weighting table is indicative — it cannot be used to predict exact scores.
"Describe Image is ~31% of Speaking, so a perfect Describe Image should give me 31% of the 80-point GSE range."
Raw score points do not translate directly into scaled points. The scoring algorithm combines your performance with the difficulty of each question and its weighting before producing the scaled Speaking score.
"Incorrect answers make you lose your points."
The lowest possible score for any question is zero. Some MCQ-Multiple-Answers tasks use partial credit: +1 for a correct option, −1 for an incorrect one — but the question total cannot go below zero. The rule discourages selecting every option, while protecting you from negative scores.
"Completing just two Summarize Written Text tasks in the Scored Practice Test gave a high Writing score, so that must be how the operational test works."
The Scored Practice Test is accurate only when taken in full and in good faith. Selectively answering a few questions distorts the result and is not how the real test works. PTE uses a sophisticated model across all question types, difficulty and weighting — completing only one question type does not give the test enough information to assess ability.
"My score changed after a rescore — this means the original score was wrong!"
Not at all. A rescore is done entirely by two trained human raters per response (with a third adjudicator if they disagree). Small differences in human judgments can accumulate, leading to a slightly different result — most rescores closely match the original. Differences do not indicate error; they reflect the inherent subtleties of full human evaluation.
"The appeal process will re-evaluate my test responses and explain why my scores changed."
PTE Academic offers three different post-results services: 1. Technical review — checks for technical issues with audio recordings. Does not change scores. 2. Human rescore — every open-ended response is re-evaluated by human experts; the new score replaces the original (it can go up, down, or stay the same). 3. Appeal — verifies that the rescore was carried out correctly per Pearson's procedures. It does NOT re-assess your responses and does not provide individual feedback.