« IELTS Writing Task 2: 'different perspectives' paragraph | Main | IELTS Speaking Part 3: four-step answer »

Thursday, September 20, 2018


Feed You can follow this conversation by subscribing to the comment feed for this post.

Thumps up! you teach us very good methods in how to write both task1 and task2. The rest is for us to take practices by using your methods.
When getting familiar with your approaches, we will be able to deliver it into test smoothly under pressure.

Thank you so much sir.This is the method I am searching for.Now,we can apply this approach in task 1 type questions.

The bar chart reports the incidence of dining out at fast food establishments in the USA in 2003, 2006 and 2013.

The results resemble a bell-shaped curve, with peak incidence in the two median groupings (once-per-week, and once-per-fortnight or month). The final results for the decade are skewed more towards a drop-off in frequency.

In all three years, the two median categories accounted for around sixty percent of the responses, slightly less than in a standard distribution. The shoulders are wider than standard, with these cohorts, representing several times per week and a few times per year, showing incidences of between twenty and thirteen percent respectively, which suggests a wider variance than normal. The tails are also fatter than normal, varying between just over two percent to five.

There was a small trend toward a higher frequency in 2006, but the final results are skewed the other way, evidencing a general decline over the decade, with the median result shifting from weekly (31%), toward once a fortnight or month (34%), with a concomitant fall in the weekly cohort to about 26%, and the upper shoulder rising to fifteen percent, while the lower shoulder dropped to sixteen percent.

(<2000 words)

Hi M'ambo,

Your level of English is obviously very good, but I worry that you are trying a little too hard here. IELTS candidates are not expected to use the kind of 'statistics language' that you employ e.g. bell-shaped curve, median groupings, skewed towards, standard distribution, shoulders, wider variance, tails are fatter, median result, upper/lower shoulder.

Personally, I wouldn't use these phrases in an IELTS report.

I also worry that your desire to write a statistics-style report takes your focus away from describing the figures in a more direct and obvious way. You're making the examiner do a lot of work to assess whether your sentences are truly accurate, or whether you're simply trying to impress with big words.

A few examples of the above:

- There is no mention of "eating in fast food restaurants" anywhere after the introduction, so you're making things difficult for the examiner.
- In paragraph 3, the examiner may think "60% of what responses?" or "what are the shoulders exactly?"
- The examiner may also think: "why are you comparing this to a standard distribution?" or "what is the normal variance?" that you refer to, and "what are the fatter tails?"

I would advise you to stick to a simpler style. Just describe and compare the figures as you see them - without referring to any 'outside' statistical knowledge - and keep everthing as clear, coherent and concise as possible.

Thank you for commenting, Mr Simon, point taken.

The bar graph illustrated the frequency of eating in fast food restaurants by Americans in 2003, 2006, and 2013.

Overall, their preference mostly remained stable over the course of 10 years. In particular, about 60% of American citizens paid one to four visits per months to fast food restaurants.

In 2003, the most common frequency of eating fast foods at restaurants was once a week, with the figure sat at around 31%. The next highest figure in 2003 was of once or twice a month, being only 1% less than the highest. Lower than that were the figures of “several times a week” and “a few times a years”. Few people chose “everyday” or “never”; none of the figures for those choices were higher than 5%. The same trend happened more or less the same with year 2006 too. The only outlier occurred in 2013, when the proportion of people eating fast foods once or twice per month was the highest (about 33%), comparing to all other choices of the same year.

Anh Tran

  • It is usually: The bar graph illustrates...
  • per month
  • with the figure sitting at
  • the next highest figure in 2003 was for the category once or twice a month
  • figures for
  • chose: perhaps "responded" on the assumption that the data comes from a survey; but better to write in terms of "incidence"
  • "Trend" refers to an average line through three or more data points spread out over time. So we could talk about the trend toward an overall lower frequency over the decade, but there cannot be a "trend" if we are looking at just one point in time.
  • compared to
  • The results for 2013 were not so much an outlier as an possible overall trend: the higher frequencies dropped and the lower frequencies increased, so the distribution as a whole changed. Whether this is really a trend or just a blip or temporary anomaly is hard to tell because we do not have the results for the intervening years. It could go either way.
  • Dear Simon,

    I love the way this super coherence thinking improves the logic of the whole report.

    I have a question regarding paragraphing.
    We are often instructed to write a separate overview paragraph. However, is it possible to write all the green parts in one single paragraph, followed by all the blue part in another paragraph? In other words, a general comment + details.
    I’m not sure if the examiners would treat this as “missing overview paragraph”.

    Looking forward to your guide.

    Thank you, Mithrandir.

    The reason why I wrote the first sentence in the past tense was to make the essay being consistent in one grammatical tense. Not sure if it is the optimal approach.

    Definitely agree with you about the usage of "trend" and "outlier".


    You could do that, but you would be making the examiner's job more difficult. He/she will be looking for an overview (because it's an important aspect of the scoring criteria), so it's best to make it obvious. Using your suggestion, the examiner has to look much more carefully.

    It seems that you did not use this 'supercohesion' method in your previous Task 1 essays(or perhaps you did). But anyway,I was still wondering if you would give me some insights into using this method.Any previous examples will be lovely. Thank you.

    I feel that super cohesion can really save us from not knowing what to write in paragraph 3 and 4, especially when dealing with diagrams that have tons of information.I was just wondering if you would mind producing some more examples or rewriting some of the previous essays using super cohesion(Sorry for recommenting and using the 'insights' term incorrectly though).

    Ji Yan

    I have rewritten Simon's answer using 'super cohesion' guidelines to this chart:


    Simon's original answer is here:

    The bar chart compares the incidence, by six age cohorts, of patients attending emergency departments of hospitals in Northern Ireland in the month of December 2016 and during the same period one year later.

    The most frequent attendees at A&E (accident and emergency) departments were either infants or the elderly. There was also an upward trend in attendances across the board.

    The highest incidence was recorded for those aged 75 or over, with a rate in the upper sixties per 1000-population. This was over double the rate for the middle cohorts (16 to 44, 45 to 64, and 65 to 74) which was around thirty per mille in both months under review.

    At the other extreme, A&E presentations by under-5s were just over fifty per 1000-population for both months: again more than double the next group up, children aged 5 to 15, who made the fewest visits, at just over 20 per 1000-population.

    The increase in emergency care attendances was around a few percent on a year-on-year basis, and roughly of the same proportion in all categories.

    (~180 words)

    Ji Yan

    Simon's answer for the maps question pretty much follows 'super cohesion' principles amyway.


    Ji Yan

    An even earlier task:



    A 'super-cohesion' version:

    The line graph compares three types of travelers visiting New Zealand between 1997 and 2017 in terms of their mean daily spend during their trips.

    Overall, spending by international visitors to New Zealand peaked between 2000 and 2003, and ended the two decades well below the earlier peaks. Spending by business travellers never recovered, lapsing into second place behind holiday and vacation spending, while spending by friends and relatives remained a distant third throughout.

    All groups rose substantially toward 2002/3, business and vacation spending by about a third, while spending by those visiting friends and relatives nearly doubled. All three groups then dropped steeply, business spending by nearly half. In the decade to 2017, business spending became erratic, holiday/vacation spending showed an upward trend, and the third category (re friends and relatives) flattened out.

    In 1997, at $260, spending by visitors on business was over twice that of those visiting friends or relatives and some $60 higher than that of vacationers. By 2017, the average for holiday/vacation spending had recovered to $250 per diem, significantly more than its 1997 value, $40 above the business figure, and was running at nearly twice the rate for friends and relatives ($140).

    (< 200 words)


    Ji Yan asked a useful question, and I'd like to thank Wanda for some good work in response!

    It's true that I didn't consciously use this super-cohesion approach until recently. Perhaps I sometimes used it without thinking.

    Now that we have a memorable name for this approach, I'm sure it will be in my mind whenever I'm writing a task 1 answer. I'm not sure whether it will always be appropriate or usable, but we'll see!

    This is super useful! Thanks, Simon.

    here is my essay:

    The bar chart illustrates the percentage of meals that American people ate in fast food restaurants in 2003, 2006 and 2013.

    Looking at the chart it is clear that most of the people used to go to fast food restaurants once a week and once or twice a month. Also, the diagram outlines a downward trend of using fast food in 2013.

    According to the chart, around 30% of American had a fast food meal once a week same as the group who had once or twice a month, while the percentage of people who avoid junk foods or in contract, ate fast food every day was below 5% for both categories. The number of people who were eating at these restaurants multiple times every week varies from 16% to 20% during these years. Also, 15% of American citizen chose this kind of food just several time during a year in 2006 and 2013.

    There is a fluctuating general trend but in 2013, except two groups that avoid fast foods or used them just for a few meals during a year, other group’s usage was decreasing to once or twice per month in comparison whit previous years.


    1) "the percentage of meals": no, this is factually incorrect. It shows the percentage of people in the USA broken down by incidence category.

    2) "trend of using fast food": this would include takeaways. The chart relates only to the customer count for fast food meals eaten on site. The same applies to "had a fast food meal".

    3) "the same as"

    4)-> who avoided junk [junk food is not the same as fast food, although the two may overlap] foods or in contrast

    5) 15% of American citizens ... several times per year

    6) ... except for the two groups that avoided fast ...

    7) " other group’s usage was decreasing to once or twice per month in comparison whit previous years": factually inaccurate. -> the remaining upper frequency groups generally showed a decline, balanced by an increase in those eating out at the lower frequency of once or twice per month.

    thank you for correcting Mr Simon

    @Wanda Hi, can you tell me if a native speaker?

    Hi @Wanda , can you tell me if you are a native speaker?

    Hi Simon,

    I have a couple of questions about your essay:
    . Is it alright for the examiners when using "from 2003 to 2013", whereas the bar chart just mentioned about 3 separated years?
    . Since the graph just showed the data in 3 years, can we assume the shift towards eating habit

    Thank you.

    Hi Simon!

    Do you think this structure would also work for this essay?

    1. Topic sentence of the whole essay (=what the graphic is about)
    2. What comparisons you'll make (comparison A and B: without actually making the comparisons) and HOW you'll make them.
    (3rd sentence - if there's a lot of information)
    Paragraph 1
    2. Topic sentence - what comparison A is
    The main idea of each paragraph (=the topic sentence) is followed by details that support it, so:
    2, 3, 4 - details of that comparison
    Paragraph 2
    2. Topic sentence - what comparison A is
    2, 3, 4 - details of that comparison
    Summary of the ideas I've developed in my essay. In the introduction, I made a general statement about the comparisons I planned to make. The conclusion is a restatement of these ideas.

    does anybody know why Simon wrote four times in the sentence'In each of the years shown on the chart, close to 60% of people in the US ate in fast food restaurants between once and four times a month. '? is this an error?

    Verify your Comment

    Previewing your Comment

    This is only a preview. Your comment has not yet been posted.

    Your comment could not be posted. Error type:
    Your comment has been posted. Post another comment

    The letters and numbers you entered did not match the image. Please try again.

    As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

    Having trouble reading this image? View an alternate.


    Post a comment

    Your Information

    (Name is required. Email address will not be displayed with the comment.)