A common assumption is that a Task 2 essay is judged as a whole, with one examiner reading it through and settling on an overall impression somewhere between 6 and 7. That is not how the number is produced. The final band is the result of four separate judgements, arithmetic that averages them, a rounding rule that can nudge the outcome up or down, and in many cases a second examiner who never sees the first one’s marks. Understanding that machinery explains why two essays that feel similar can land on different bands.
The four descriptors that split your essay into quarters
Every Task 2 response is scored against four criteria, and each one carries equal weight. Task Response looks at whether you actually answered the question asked, developed a position, and supported it. Coherence and Cohesion covers the logical flow of ideas and how sentences and paragraphs connect. Lexical Resource assesses the range and accuracy of your vocabulary. Grammatical Range and Accuracy examines the variety of your sentence structures and how many errors slip through.
The examiner assigns a whole-number band from 0 to 9 for each of these four areas independently. A strong argument does not lift your grammar score, and a wide vocabulary does not rescue a weak answer to the actual prompt. Each descriptor is marked on its own descriptor scale, using detailed public and examiner-only band descriptors that spell out what a 6 looks like versus a 7.
How the four sub-scores are averaged into one figure
Once the four band scores exist, they are simply added together and divided by four. If you score 7 for Task Response, 6 for Coherence and Cohesion, 7 for Lexical Resource, and 6 for Grammatical Range and Accuracy, the total is 26, and 26 divided by 4 is 6.5. That average becomes your Task 2 band before it is combined with Task 1 into a single Writing band.
Because the average is unweighted, one weak criterion drags the whole essay down by a quarter of its shortfall. Dropping a single descriptor from 7 to 6 removes 0.25 from the average. That is why a candidate cannot afford to neglect any one of the four areas in the hope that the others will compensate.
Where half-bands and rounding rules change your result
The average of four whole numbers does not always land on a tidy half. Four sub-scores can average to values like 6.25 or 6.75, and Writing bands are only reported in whole and half bands. This is where the rounding rule matters. An average ending in .25 is rounded up to the next half band, and an average ending in .75 is rounded up to the next whole band.
So sub-scores of 7, 6, 6, 6 average to 6.25 and are reported as 6.5. Sub-scores of 7, 7, 7, 6 average to 6.75 and round to 7. The rounding always works in the candidate’s favour at these two points, but it also means the difference between a reported 6.5 and a 7 can come down to a single descriptor moving by one band. Small, consistent improvements across all four criteria are often what tips an average past a rounding threshold.
The double-marking check that catches inconsistent scores
Writing scripts are frequently marked by more than one examiner, and the second marker does not see the first marker’s decisions. When two independently produced sets of band scores disagree by more than a set tolerance, the script is referred to a senior examiner who resolves the discrepancy. This is the quality control that stops a single harsh or lenient reading from deciding your result.
That same principle of independent re-checking is why a formal review can change an outcome. If your reported band feels out of step with your performance, comparing an Enquiry on Results against a single-skill retake, as explained at https://careerwiseenglish.com.au/, helps you decide whether a re-mark or a fresh sitting is the more sensible route. A re-mark puts your existing script in front of a senior examiner; a retake gives you a new script to be double-marked from scratch.
Knowing the arithmetic is only useful if you keep practising against all four descriptors rather than polishing the one you already do well. Revisit your weakest criterion regularly, because that is the number most likely to be holding your average just below the next rounding point.