# To investigate the difference between broadsheet and tabloid newspapers in terms of the language they use.

Extracts from this document...

Introduction

Statistics Coursework

## Task

To investigate the difference between broadsheet and tabloid newspapers in terms of the language they use.

## Hypothesis

The hypothesis I am going to investigate is:

Broadsheet newspapers use longer sentences, paragraphs and articles than tabloid newspapers. Also the language they use is that of a higher reading level than that of tabloids.

## Method:

## Data Collection

To do this I have taken a common article across an equal sample of tabloid and broadsheet newspaper for the same day and analysed the text. I am using a common article so that the material covered is similar and the writing style is likely to be comparable and this therefore removes a variable.

My sample uses three popular tabloids: The Sun, The Daily Express and The Daily Mail and three popular broadsheets: The Guardian, The Times and The Independent. I chose to do three of each type due to limited time and money available. To select the newspapers I used a systematic sampling method of writing down a list of all the daily newspapers and selecting every third one.

The data I collected from each newspaper was:

- The number of words in each sentence throughout the whole article
- The number of sentences in each paragraph throughout the whole article
- The total number of words in the whole article
- The total number of sentences in the whole article
- The total number of paragraphs in the whole article
- The number of words with three or more syllables in the first 10 sentences

I have collected this data in tally charts and stem and leaf diagrams.

I used tally charts for the sentences per paragraph, as it was the easiest method for counting and displaying the data. From this it was simple to create a frequency table. See Table 1.

Middle

When we compare the two box and whisker plots (see Graph 4) for the total types we can see that the interquartile range of both are very similar however the overall range of the tabloid is a lot smaller that that of broadsheet. This shows that broadsheet newspapers have more long sentences than tabloids.

The standard deviation also shows the spread, but by its distance from the mean. The bigger the standard deviation, the bigger the spread of values from the mean, hence the less consistent the data. The standard deviation is calculated by the formula on page 3 and recorded on Table 4.

I have calculated the percentage of the data that is within one standard deviation of the mean below. If the percentage is 68, then it is a normal distribution (and will form a bell shaped frequency curve). If the data is above that it is positively skewed and if it is below it is negatively skewed.

Broadsheets –

s.d. = 11.409 mean = 24.454

24.454 + 11.409 = 35.863 24.454 – 11.409 = 13.045

79 out of 130 sentences have within 13.045 to 35.863 words (i.e. 14 to 35)

Therefore 79 x 100 = 60.8% of the data is within 1 standard deviation of the

130

mean.

#### Tabloids -

s.d. = 8.462 mean = 23.416

23.416 + 8.462 = 31.878 23.416 – 8.462 = 14.954

59 out of 89 sentences have within 14.954 to 31.878 words. (i.e. 15 to 31)

Therefore 59 x 100 = 66.3% of the data is within 1 standard deviation of the

89

mean.

These are both less than the normal distribution of 68% so the graphs they would produce are negatively skewed.

I now know that these are skewed but using Pearson’s coefficient of skewness I can calculate just how skewed they are. This skew is illustrated on Graph 6 and 7.

Skewness = mean – mode

standard deviation

Broadsheet -

24.454 – 31 = - 0.57

11.409

Tabloid -

23.416 – 29 = - 0.66

8.462

#### Averages

There are three different types of average. Mean, median and mode. Their definitions are on page 2. I have calculated each of these averages. See Table 4

I have plotted these calculations on Graph 6 and 7.

Conclusion

The data collection was a very lengthy process as every word, sentence and paragraph had to be counted, this was part of the reason I only covered 6 newspapers. It would have been desirable to have looked at a wider range of newspapers, as it would have given a more apparent conclusion; and look at letters or syllables per word, as I think it would give a clearer indication of the difference uses of language between broadsheets and tabloids. It would be preferable if the data collection process could be carried out perhaps automatically by computer to cut down the time it takes.

I found that, especially for sentences per paragraph, it was fairly useless calculating the mode and the median as they do not provide the ‘typical’ value and therefore the mean was much more useful in analysis.

My investigation was far less straightforward than that of other pupils and hence it was quite difficult to find statistical analyses that could be carried out!

I was quite surprised to find that there was not much published secondary data on newspaper comparisons. I was expecting to find lots on the Internet but none was found. As you can see I had to settle for data that was not wholly relevant as it was all I could find. This meant that the secondary data did not back up my data very well.

If I was to re-do or further the investigation, I would collect the required information using a computer, collect data on word length and perhaps syllables per word, and collect data from more sources. I could also analyse this data not only for serious articles but for something light-hearted such as sport or reviews as well. It would also perhaps be a possibility to see how the conclusion I obtain relates to sales of each newspaper and who the actual audience is.

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