• Join over 1.2 million students every month
  • Accelerate your learning by 29%
  • Unlimited access from just £6.99 per month
  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10
  • Level: GCSE
  • Subject: Maths
  • Word count: 4017

Compare mass-appeal tabloid newspapers and quality newspapers by attempting to find statistical differences.

Extracts from this document...


GCSE Statistics Coursework

Aims of the project

        I aim to compare mass-appeal tabloid newspapers and quality newspapers by attempting to find statistical differences. To represent the mass-appeal papers, I chose the Daily Mirror and for the text-quality based newspapers, I chose the Times. Hopefully, there will be some significant statistical differences in the style of journalism which I will be able to comment on.


Data Collection: I decided to choose similar pages from both the Times and the Mirror with roughly equal numbers of paragraphs and adverts, pages 4-5, or 4-6, as in the Mirror there were not enough sentences to take samples from. To find mean sentence lengths in the two papers, I decided I would sample systematically from my populations, counting the number of words in every 3rd sentence. I came up on several problems quickly – should I include headlines in my count? I decided against it, as headlines tend to be shorter than normal sentences. The next problem came with numbers – did they get counted as words in the sentences? Making sure that I did the same with both papers, I decided to exclude numbers in my count. I also decided to exclude any sentences in adverts, as the number of adverts on the compared pages varied. I then took a mean and found the standard deviation of my data.

        To find the average number of words per sentence, I decided to ‘cluster-sample’, and count the first 30 words in the first paragraph of page 4 in each paper. I decided that I would again exclude numbers, and that hyphenated words counted as a single word. Again, when I found all the data, I found its mean and the standard deviation.


...read more.











*Please note that all approximate difference in this paper are calculated to be 100{(1±T/M)+(1±M/T)}/2 rounded to the nearest 5% when T=figure for Times and M=figure for Mirror.

The reason why I drew a variable width histogram to compare the two statistics is that if I had drawn a box-and-a-whisker diagram, the outliers, in both papers, would have been excluded in both papers, but I want the outliers to be included because I think they do have some reflection on the style of journalism. Also, histograms show well the general distribution.

Interpretation of Results

Location: The top histogram shows the sentence lengths in the Times and the bottom one the Mirror. The mean sentence lengths in Times (26) is substantially higher than that of the Mirror (19), by nearly 8 words per sentence. When the means in both papers do not exceed even 30, this is quite something; 35% difference is significant, I would say.

Spread: The standard deviation for Times is 11, and for Mirror it is 9, again showing that Times has a wider spread, although this is only slight. The 20% difference in both standard deviation and IQR indicates a greater variety of journalism in the Times. Not only that, the Mirror has less sentences anyway so it would be harder to find sentences which would be considered as outliers in that they are longer than the ‘average’ sentence.

Skew: The sentence lengths in both papers seem to be positively skewed, though in the Times, it is very slight, co gay pared with the Mirror. Naturally, you would expect there to be positive skew. However, I do not think there is much to be commented on this.

For further investigation, I could to the normal distribution test, but I do not have to.

Conclusion: There was a quite significant difference in the sentence lengths. I reject H(0) in favour of H(1); the Times has quite significantly longer sentences than the Mirror. The fact that the Times has longer sentences could be due to a number of reasons. The Mirror is a shorter paper than the Times; so a way to deal with as many subjects and still keep ‘thin’ is to cut down on the provision of tiny details to the readers. In the Times, small details usually tend to be crammed in the same sentence after the main clause, effectively lengthening the sentences.

The Times did have slightly larger spread of sentence lengths, but not so much that I can say the difference was significant. However, the spread occurs at different levels; the median of the Times’ sentence lengths is 27, whereas in Mirror it was 19.

Hypothesis 2

-        H(0) The two newspapers will have similar word lengths.

      H(1) Times will have longer word lengths.

Diagram: I have chosen to use a box plot to compare the word lengths of the two papers because they show clearly the location, spread and skew of the data at one glance. image07.jpg

Location: The median in both cases lie on 4. The differences in the mean between the two papers is under 0.5 letters, with the Times being 4.7 and the Mirror 4.3, which is quite close, and renders itself insignificant.

Spread: The spread in the Times seems to be larger. Its IQ range is 4, whereas in the Mirror it is 2. You could comment that this does show a larger variety in the vocabulary used in the Times, but it might just be that there happened to be less prepositions in the Times sample, and as the sample was taken systematically, this could may well be the case.

Skew: The results in both papers are positively skewed. However, one would naturally expect that because of the presence of prepositions and articles in the English language, which occur very frequently and usually are very short, often only one or two letters.

Further Investigation: Seeing that there was no significant difference in the world lengths of the two papers, I wondered if this is so in all English literatures. So, I did a test to see if this was the case. I took 20 words from two books aimed at different age groups. The sample was selected by choosing every 4th word from a random page. I chose a book that is read by 9~10 year olds and another which has is aimed at adults. I found that the results are much the same as the results for the two newspapers. They have the same median to start with; and they share the lower quartile. However, ‘The Glass Bead Game’ has a wider IQR, indicating that its spread in lengths of the words is bigger than that of the children’s book.


Conclusion: There is no significant difference. H(1) is rejected in favour of H(0). There was actually more difference in the pre-test samples. Maybe this indifference in word lengths is something that is present in all English prose, as the further test has shown. It seems that there is not much to say about the comparison between the two papers in terms of word lengths. The actual word lengths are so similar that I cannot conclude much from them; their median was the same, for example.  Although the IQR in the Times was bigger than that of the Mirror by 2, I think that was just ‘unlucky’; with the IQR being so small and the figures heavily clustered around the median, it could just be by chance that the there happened to be one or two less 2- or 3-lettered words.

Hypothesis 3

  • H(0) There will be no significant difference in the text area to rest area ratio in the pages of the two newspapers.

H(1) There will be a significant difference in the text area to rest area ratio in the pages of the two papers.


Diagram: I have drawn two scatter diagrams, with the y-axis being the percentage of the non-text area in the pages and the x-axis the percentage of text area in the corresponding pages. This means that the sum of the x co-ordinate and the y co-ordinate of a point always adds up to 100. So, all the points in the scatter diagrams lie on a straight diagonal line between (0,100) and (100,0) on the graph. However, it is not the gradient that matters in these two diagrams, as they both have the same; it is the location, spread and skew of the co-ordinates, which will give us information about the trend in the text area to rest area in the newspapers.image10.png

        The sample size is not too big, as I only took the data from 8 pages. Still, I hope that what little sample I have would show some significant differences.image01.png




Approx. difference

Mean percentage of

Non-text area



40% (20% if you just take 63-43)

Standard deviation of

Non-text area




...read more.


Unfortunately, the second hypothesis turned out to be bit of a ‘dud’ and nothing much can be indicated in terms of word lengths.

However, I think that the third and the fourth hypotheses say more than the sentence lengths about the general editorial and the style of journalism in the two newspapers.

It is clear that there is less room for text in the Mirror, because so much is taken up by pictures. 17.5 is quite something, and as I have said before, that is not counting the headline areas. I have seen pages in the Mirror where the headlines occupy almost a half of the paper. Also, the picture to text area ratio in the Times is more consistent.

The articles in the Mirror are also shorter, as I have proved above, which is in accordance with the fact that there is less room for the text anyway. So why do people bother buying the Mirror then? The truth is, I must conclude, that the Mirror, as with all mass-appeal tabloids, sells itself by the pictures, ‘page 3 girls’ and catchy headlines. Therefore, not many people who read the Mirror care about its quality or the style of journalism, because more often than not, they will just skim over the text and look and the pictures. Also, the Mirror come up with many ridiculous stories and people are just interested in what happens because what happens is so outrageous. On the other hand, the readers of the Times want accurate information, and a lot of it.

People READ the Times and LOOK at the Mirror.

...read more.

This student written piece of work is one of many that can be found in our GCSE Comparing length of words in newspapers section.

Found what you're looking for?

  • Start learning 29% faster today
  • 150,000+ documents available
  • Just £6.99 a month

Not the one? Search for your essay title...
  • Join over 1.2 million students every month
  • Accelerate your learning by 29%
  • Unlimited access from just £6.99 per month

See related essaysSee related essays

Related GCSE Comparing length of words in newspapers essays

  1. Write a hypothesis about the length of words in newspapers and magazines.

    To extend the current investigation using the same selected articles, it has been decided to focus on the number of words in one sentence. The extension will follow the same format. 7.0 Extension The previous investigation proved my stated hypothesis to be incorrect.

  2. "Broadsheet newspapers have a longer average word length than tabloid newspapers"

    of letters in word Tally Frequency 1 2 3 4 5 6 7 8 9 10 11 12 13 14 llll llll llll llll llll ll llll llll llll llll l llll llll lll llll llll llll llll lll llll lll llll llll l l 5 22 21 13 4

  1. Outline any differences between Tabloid and Broadsheet Newspapers in terms of word length, sentence ...

    100 of Text (%) Total area of page (cm�) Another way of working out the percentage of text on the page would be by working out the area devoted to advertisements and pictures using the above method (to calculate area of text).

  2. Maths Statistical Coursework

    in the form of pie charts, as I believe that they are easily comparable by simply looking at the pie charts, of course the topics of interest in this case are politics, global issues and celebrity/personal stories. Daily Mail Daily Star The Independent Observations As a result of the calculations

  1. GCSE Statistics Coursework

    3 7 8 4 5 13 5 5 18 6 11 29 The 21.5th number lies in this category so this means that the median is six for the Daily Mirror. Upper Quartile = 42 [Total amount of Numbers (n)] + 1 / 4 = 9.75 x 3 = 32.25th number.

  2. Statistically comparing books

    Nicholas Nickleby Number of words Frequency (f) Mid point (x) fx fx^2 Cumulative frequency 0<x>5 10 2.5 25 62.5 10 5<x>10 13 7.5 97.5 731.25 23 10<x>20 14 15 210 3150 37 20<x>30 8 25 200 5000 45 30<x>50 14 40 560 22400 59 50<x>75 8 62.5 500 31250 67

  1. Maths Statistics Coursework

    This showed that the broadsheet contained more of the shorter words than the tabloid and the tabloid contained more of the longer words.

  2. GCSE Maths Handling Data Coursework - Comparing Newspapers

    If we divide a cumulative frequency curve into quarters, the value at the lower quarter is referred to as the lower quartile, the value at the middle gives the median and the value at the upper quarter is the upper quartile.

  • Over 160,000 pieces
    of student written work
  • Annotated by
    experienced teachers
  • Ideas and feedback to
    improve your own work