Task 1
You should spend about 20 minutes on this task.
The bar chart below shows the number of people employed in three sectors in one country in 1990 and 2020, and the line graph shows total employment in the same country between 1990 and 2020.
Summarise the information by selecting and reporting the main features, and make comparisons where relevant.
People employed by sector, 1990 and 2020
| Agriculture | Manufacturing | Services | |
|---|---|---|---|
| 1990 | 6.2 | 8.4 | 11.9 |
| 2020 | 2.1 | 5.3 | 24.6 |
Total employment, 1990–2020
| 1990 | 1995 | 2000 | 2005 | 2009 | 2012 | 2015 | 2020 | |
|---|---|---|---|---|---|---|---|---|
| Total employment | 26.5 | 27.3 | 29 | 30.6 | 28.2 | 29.4 | 30.9 | 32 |
Write at least 150 words.
Task 2
You should spend about 40 minutes on this task.
Some people argue that the main purpose of prisons should be to rehabilitate offenders so that they can return to society as law-abiding citizens, rather than to punish them for the crimes they have committed.
To what extent do you agree or disagree?
Give reasons for your answer and include any relevant examples from your own knowledge or experience.
Write at least 250 words.
What a Band 8–9 response must demonstrate
Measured against the four official assessment criteria — specific to this test's two tasks.
Task Achievement
- Gives a clear overview that links the two charts: total employment grew over the period despite a dip around 2009, and that growth was driven entirely by services, while agriculture and manufacturing both shrank.
- Selects the key features rather than reporting every figure: services roughly doubling (11.9 to 24.6 million), agriculture falling by about two-thirds, and the 2009 trough on the line graph.
- Supports the main features with accurate figures and uses them to compare — e.g. services employed fewer than twice as many as manufacturing in 1990 but more than four times as many by 2020.
- Stays descriptive: no speculation about recessions, deindustrialisation or policy.
Coherence & Cohesion
- Organises the report around the data, not the charts in isolation — typically an introduction, an overview covering both charts, a paragraph on the sector shifts and a paragraph on the total-employment trend with its dip.
- Uses referencing ("the former", "this sector", "over the same thirty-year period") to move between the bar chart and the line graph without restating titles each time.
- Cohesive devices signal the relationship between the two charts ("despite this", "meanwhile", "in line with") instead of being stacked mechanically.
Lexical Resource
- Uses precise trend and magnitude language: employment in services more than doubled; agriculture contracted sharply; total employment dipped before recovering.
- Varies the vocabulary for people in work: employed, workforce, jobs, those working in, the number in employment — and avoids repeating "people employed" in every sentence.
- Handles the unit correctly as millions of people, not percentages or shares.
Grammatical Range & Accuracy
- Controls comparative and multiplicative structures ("almost four times as many", "a third of the 1990 figure") and participle phrases ("having fallen to 28.2 million in 2009, total employment…").
- Uses past simple consistently for the dated period, with "by 2020" and "between 2005 and 2009" time frames handled accurately.
- Writes figures and units consistently (24.6 million, not "24,6m"); the majority of sentences are error-free.
Task Response
- Takes a clear position on the extent of agreement — for instance, agreeing that rehabilitation should be the primary aim while arguing that it cannot entirely replace punishment — and states it in the introduction.
- Addresses the word main: the question is about priority, not about whether rehabilitation is good. Essays that simply praise rehabilitation without weighing it against punishment do not fully answer the task.
- Develops ideas with specific support: reoffending rates, vocational or education programmes in prison, the deterrent and justice functions of a sentence, the effect on victims.
- Sustains the position through to the conclusion rather than drifting into a balanced "both matter" ending that contradicts the introduction.
Coherence & Cohesion
- A four- or five-paragraph structure with one controlling idea per body paragraph and topic sentences that signal it.
- Logical progression: the essay builds a case (why rehabilitation reduces crime, what punishment still provides, how the two can be weighted) rather than listing unrelated points.
- Cohesion through referencing and sequencing, not a linker at the start of every sentence.
Lexical Resource
- Uses precise, topic-appropriate vocabulary: reoffending, recidivism, deterrent, retribution, custodial sentence, reintegration, parole, vocational training.
- Paraphrases the prompt rather than copying it ("rehabilitate offenders" → "help those convicted of crimes to change their behaviour").
- Uses collocations naturally ("serve a sentence", "a revolving door", "pay one's debt to society") without forcing rare words.
Grammatical Range & Accuracy
- A controlled mix of complex structures: conditionals ("If prisons did nothing more than confine people…"), concession ("While punishment satisfies a sense of justice…"), and relative clauses.
- Accurate modal use for hedged argument ("is likely to", "may well", "should"), avoiding absolute claims such as "prison never works".
- Punctuation supports complex sentences with no comma splices; errors are rare and do not impede meaning.
Common traps for this specific test
The mistakes most likely to cost marks on these two questions.
The charts tell one story: total employment rose because services grew faster than the other two sectors shrank. A report that treats the bar chart and the line graph as unrelated, with no sentence connecting them, loses Task Achievement marks for a weak overview.
Total employment does not rise smoothly. It falls from 30.6 million in 2005 to 28.2 million in 2009 before recovering. Writing that employment "increased steadily throughout the period" misreads the graph; the trough is a key feature.
The three sectors in 1990 add up to 26.5 million, matching the line graph, and 32.0 million in 2020. Candidates sometimes treat the bar chart as percentages or invent extra sectors; read the unit (millions of people) carefully.
Eight data points do not all need reporting. A Band 8 answer gives the start, the 2005 high, the 2009 low and the 2020 end, and summarises the rest as a general upward trend.
Attributing the 2009 dip to a financial crisis or the decline in manufacturing to offshoring is speculation and is not rewarded in Task 1. Describe, compare, and stop.
The question is about priority. An essay that lists the benefits of rehabilitation but never weighs it against punishment, deterrence or justice for victims has only half-addressed the prompt.
"To what extent" still requires a position. "Both are important" with no indication of which should take precedence is not a clear view and caps Task Response at Band 6.
Statements such as "everyone who goes to prison reoffends" or "punishment never works" are unsupported absolutes. Hedged, specific claims ("reoffending rates tend to be lower where prisoners gain qualifications") score higher for both Task Response and Grammatical Range.