Looking at graph two you can see that the trend of data for each paper is almost identical which shows that there is not much difference between each paper. The graph does show however that the broadsheet does contain more longer words than the tabloid.
I have come to the question are the words used in a political article longer than those used in a sport article so I have decided to extend my original investigation.
Extended Hypothesis
I expect that the word length in a political article to be longer than that of a sport article.
I repeated the process that I had done for my political article and my results were:
Daily Telegraph (broadsheet Birmingham Vs Arsenal
Daily Mail (Tabloid Birmingham Vs Arsenal)
Daily Telegraph
Mean: 515/100 = 5.15
Median: 5
Mode: 3.5
Range: 11
Daily Mail
Mean: 377/100 = 3.77
Median: 4
Mode: 4
Range: 12
To compare these results against each other I produced another set of graphs.
Looking at my bar chats of this data (graph three) you can see that the data is almost identical with both articles containing words with 13 letters. This could be due to the fact of the names of the sports teams such as Birmingham which would make up for the extra long words in both articles than that compared to the political articles. Looking at my frequency polygon (graph four) again we can see that the trends of these sets of data are almost identical.
To compare all the data I have decided to show it in the form of cumulative frequency charts. The cumulative frequency was worked out for all 4 articles used and here are the results.
Cumulative Frequency for Daily Telegraph article on the Iraq War (Broadsheet Political)
Cumulative Frequency for Mirror Article on the Iraq War (Tabloid Political)
Cumulative Frequency for Daily Telegraph on the match Birmingham Vs Arsenal (Broadsheet Sport)
Cumulative Frequency for Daily Mail on the match Birmingham Vs Arsenal (Tabloid Sport)
Looking at my graph produced you can see that the trends of all 4 articles is almost identical. All four chats were put onto one set of axis to make them easier to compare and contrast. I have decided to use box plots (Fig 6) to show my median Upper and lower quartile and the inter-quartile range for each article as a box plots will make it easier to read. It would be difficult to try and read from the cumulative frequency diagram.
Looking at my box plots we can see that for the:
Broadsheet (political) Lower Quartile: 3.4
Upper Quartile: 7.3
Inter-Quartile Range: 3.9
Median: 5.1
Tabloid (Political)Lower Quartile: 3.5
Upper Quartile: 6.4
Inter-Quartile Range: 2.9
Median: 4.7
Broadsheet (Sports)Lower Quartile: 3.7
Upper Quartile: 7
Inter-Quartile Range: 3.3
Median: 4.9
Tabloid (Sports) Lower Quartile: 3.3
Upper Quartile: 6
Inter-Quartile Range: 2.7
Median: 4.4
The inter-quartile range is used to show the spread of data.
My prediction is inconclusive. The conclusion the results suggest that Broadsheets have more longer words on average than Tabloids. My evidence shows that no conclusion can be shown for the hypothesis “I expect that the word length in a political article to be longer than that of a sport article.” For the data is to close together.
Word size might not mean that the paper is more profound or easier to read, this assumption I made may be false and invalidate the investigation, I have decided to investigate further into this and see if the type of paper does effect who reads it.
Hypothesis Two
Educated people read Broadsheets.
To investigate my hypothesis I have produced a questionnaire to ask people which job they hold and which paper they read and also which level of education they went up too. I have tried to avoid biased questions so that the survey is as fair as possible the survey was from a sample of 30 people.
(For questionnaire see Fig Seven)
Prediction
I predict that more educated people will read broad sheets because of their job or what standard of qualifications they have.
My results were as follows:
The way in which I figured out if the people were uneducated or educated is if they got no higher than AS-level then they were un-educated compared to the rest of the people asked.
I tallied my results in a table:
I produced a Pie Chart to show my results (Fig eight). Looking at my pie chart I can clearly see that more educated people do read broadsheets 41%. Which proves my prediction and hypothesis correct.
Now that I have seen that more educated people I want to investigate whether or not broadsheets are harder to read than tabloids.
Hypothesis Three
Broadsheets will have a higher readably level than a tabloid.
To investigate my hypothesis I am going to use something called the SMOG readability formula. This formula uses count of words and sentence length. The formula has been devised statistically to predict readability scores. To calculate the SMOG reading grade level, you have to begin with the entire newspaper article.
If article has 30 sentences or more:
- Count off 10 consecutive sentences near the beginning, in the middle and near the end. From this sample of 30 sentences, circle off all the words containing three or more syllables (polysyllabic), including repetitions of the same word, and total the number of words circled
- Count the number of polysyllabic words in the chain of 30 sentences and look up the approximate grade level on the chart.
If Article contains less than 30 sentences:
- Count all of the polysyllabic words in the article
- Count the number of sentences
- Find the average number of polysyllabic words per sentence as follows:
- Multiply that average by the number of sentences short of 30
- Add that figure to total number of polysyllabic words and look up the approximate grade level on the chart.
Smog Conversion Table
My Results were as follows
Broadsheet (Political) the Iraq War
Had 31 sentences and 50 polysyllabic words which gives a reading level of 10
Tabloid (Political) the Iraq War
Had 15 sentences and 37 polysyllabic words
Broadsheet (Sports) Birmingham Vs Arsenal
Had 30 sentences and 64 polysyllabic words and so gives a reading level of 11
Tabloid (Sports) Birmingham Vs Arsenal
Had 32 sentences and 68 polysyllabic words which gives a reading level of 11
Looking at my results I can see that this proves my hypothesis wrong because the tabloid for the sports had a higher reading level than that for the political but this could have been due to the long words such as Birmingham that are within the sports articles.
