I will now investigate which type of newspaper has more letters per word. In order to do this, I will be selecting three articles from each of the newspapers and I will analyse and compare them. The articles I have chosen were from the same day to ensure that there will be similar articles and the articles were on the same topic.
I have decided that I will use forty words from each article and I will choose these words by using systematic sampling. I will select every fifth word in each article until I have reached my target of selecting forty words.
After selecting the words, I will make a tally of the amount of letters in each word and using these results I will make a bar chart comparing the results and I will analyse them.
When selecting my words using systematic sampling, I will not be including numbers in their numerical form (e.g. 1, 2, 3) but only if they are in their proper English form (e.g. one, two, three). This is because numbers in their numerical form are not actually words and so will not be counted.
I will also not be including the authors name, words in captions or the words in the headline as none of these are actually part of the main text which people read.
I have also decided that words joined together with a hyphen will be counted as two separate words as people readers would read these words separately in their minds as well.
The first articles which I will compare are all about Buddhist monks protesting in Burma. The Sun article is headed ‘War And Peace’, the Daily Mail article is headed ‘The Saffron Revolution’ and the Evening Standard article is headed ‘Troops Raid Monasteries To Seize Protest Leaders’
Below are the tally charts for my results from The Sun, the Daily Mail and the Evening Standard in their respective order.
The Evening Standard newspaper has a higher number of total letters than The Sun and the Daily Mail which means that the Evening Standard will have a higher mean of letters per word. This follows the prediction which I made at the beginning.
To help me understand the results more clearly, I will find out the median, mean and mode of the amount of letters per word. I predict that the Evening Standard would have a higher value than The Sun and the Daily Mail in each of these.
The median is found by writing the numbers out in ascending order of length and, in this case, finding the mean of the number of letters in the twentieth and twenty-first words.
In the Evening Standard newspaper, both the twentieth and the twenty-first words have a word length of six letters. I can see this clearly in my table as the cumulative frequency column shows me that all the words between sixteen and twenty-three have six letters. The mean of these two numbers is obviously six which give the Evening Standard newspaper a median of six letters.
In the Daily Mail, both the twentieth and twenty-first words are five letters long, giving a median of five. Also, in The Sun, the twentieth and twenty-first words are five letters long, giving a median of five.
The mean for the amount of letters in a word is worked out by adding up the total amount of letters from all the words in a sample and dividing it by the total number of words.
Due to my table, I can see that the total number of letters in my sample of the Evening Standard newspaper is 251. This is divided by forty which is the total number of words in the sample to get 6.275. This means that the mean for the Evening Standard is 6.275.
In my sample of words from the Daily Mail newspaper, there is a total of 213 letters. This is divided by forty to give a mean of 5.325. In The Sun, the total number of letters is 202 which is divided by forty to give a mean of 5.05.
The mode is the length of word which has the highest frequency.
By looking at my tally, I can see that in the Evening Standard newspaper sample, there are more six letters words then any other length of word.
In the Daily Mail, the most frequent length of word was five and also in The Sun the most frequent length of word was five.
In order to make these analyses clearer and easier to view, I have put these figures into a table shown below.
*These numbers have been rounded up to the nearest integer because you cannot have a fraction of a letter in a word.
The table shows us that the Evening Standard has a higher median, mean and mode which show that the Evening Standard has longer words in this set of articles. This also proves my recent prediction to be correct.
Below is a bar chart comparing the length of words in the Evening Standard, the Daily Mail and The Sun.
The chart shows us that the Evening Standard newspaper has a higher skew than that of The Sun and the Daily Mail. This means that in these articles, the Evening Standard newspaper has on average, more letters per word.
But, this alone is not enough to prove that the Evening Standard has longer words than The Sun and the Daily Mail as it is just the first comparison and the results may vary in the other two comparisons to come.
I will also be working out the range, the interquartile range and the standard deviation for each of these articles.
The range is found out by subtracting the lowest value from the highest value. In this case, the Evening Standard has a lowest word length of one and a highest word length of thirteen giving a range of twelve.
The Daily Mail has a range of eight and The Sun has a range of twelve.
Below I have created a cumulative frequency graph showing the cumulative frequencies of my results.
From this graph I can tell that the interquartile range of each newspaper is.
IQR of Evening Standard is 3.6
IQR of Daily Mail is 3.1
IQR of The Sun is 4
Using the box plots, I can make probability statements such as;
75% of the words in the evening standard have a word length of four or higher.
To work out the standard deviation for each newspaper I will be using the formula below.
x stands for the number of letters in a word.
f stands for the frequency of x.
N stands for the sum of the frequencies which is always forty.
x-bar stands for the mean of x.
To work out the standard deviation, I have made a table.
Using the formula, I will work out the standard deviation for The Sun. I have decided to work out the standard deviation to 2 decimal places.
√ (1238/40 – (202/40)2)
√ (30.95 – 5.052)
√ (30.95 – 25.5025)
√ 5.4575
2.34
The standard deviation for Daily Mail is 1.97 and the standard deviation for Evening Standard is 2.95.
The Evening Standard has the highest standard deviation and therefore the highest spread.
Next part is newspaper comparison part 2/3
