My second hypothesis is that the average sentence length in a broadsheet will be longer than in a tabloid, I believe this because the most complicated information would be found in the broadsheet and therefore the use connectives
Read All About It
For my maths coursework I have chosen to compare a tabloid, which is a light-hearted gossip newspaper with more celebrity banter than factual information, with a broadsheet, which contains lots of information that concerns our every day lives and less gossip, it can be described as a more 'mature' newspaper. Broadsheets are normally printed in black and white whereas tabloids are printed in colour. The two newspapers are very different and there is a big boundary between fact and fiction, which separates the two. For my coursework I have decided to use the 'mirror; as my tabloid and the 'times' as my broadsheet. I have chosen these because in themselves they are very different and it will be very interesting to see how they both compare mathematically. From my tests I hope to find out how the lengths of words and sentences both compare and also how different the language is that they both use. I predict the language each paper uses will be very different as each paper has a different purpose, for example, a broadsheet is designed to inform whereas a tabloid is designed to entertain. I also believe that there will be longer words in a broadsheet when compared to a tabloid because of this, which brings me onto my hypotheses.
Hypothesis 1.
My first hypothesis is that the average word length in a broadsheet will be longer than in a tabloid, this is because a broadsheet is intended for a more educated audience who want to be informed about the daily issues therefore they would be more impressed with the use of long, complex words, whereas the audience a Tabloid is designed for would be more impressed with the quality of the information rather than the way in which it is presented.
Hypothesis 2
My second hypothesis is that the average sentence length in a broadsheet will be longer than in a tabloid, I believe this because the most complicated information would be found in the broadsheet and therefore the use connectives and punctuation will be more frequent in a broadsheet, whereas in a tabloid the editors rely on short, snappy sentences to grab the audiences attention.
In hypothesis one the range and mean word length will be smaller in a tabloid.
In hypothesis two the range and mean sentence length will be bigger in a broadsheet.
To test my hypotheses I will use a sample of 100 words. I will collect this sample using 'systematic sampling', this means I will count the amount of words in an article and divide it by 100, this will tell me the nth term, which I can use to collect my sample. I have chosen not to use 'random sampling' and 'stratified sampling' as they would ...
This is a preview of the whole essay
In hypothesis one the range and mean word length will be smaller in a tabloid.
In hypothesis two the range and mean sentence length will be bigger in a broadsheet.
To test my hypotheses I will use a sample of 100 words. I will collect this sample using 'systematic sampling', this means I will count the amount of words in an article and divide it by 100, this will tell me the nth term, which I can use to collect my sample. I have chosen not to use 'random sampling' and 'stratified sampling' as they would take to long. To make this a fair test and avoid any bias I will take my sample from newspapers on the same day about the same event. Also. I have chosen NOT to include names of people or places, numbers (including dates), hyphens and headlines.
Word Length
Focusing on word length this is what I found for my tabloid.
No of Letters
Tally
Frequency
Fx
Cf
2
6
32
7
3
9
57
36
4
21
84
57
5
5
75
72
6
0
60
82
7
6
42
88
8
8
28
96
9
9
97
0
2
20
99
1
1
00
Mean= 5.19
Median= 4
Mode= 4
Range= 10
And these are the results for my broadsheet.
No of Letters
Tally
Frequency
Fx
Cf
3
3
3
2
4
28
7
3
24
72
41
4
9
76
60
5
0
50
70
6
9
54
79
7
5
35
84
8
7
56
91
9
7
63
98
0
0
-
98
1
0
-
98
Mean= 4.37
Median= 4
Mode= 3
Range= 8
From these tally charts I noticed that my hypothesis might not be true, as unexpectedly there seemed to be longer words in my Tabloid, this seemed very odd and I decided that I might have made a mistake. I went back and checked everything but I found nothing unordinary. I decided to put the information into a chart to be sure.
After I had put all of the information into a tally chart I worked out the mean of each graph. I did this by multiplying the number of letters by the frequency. I then added it all up and divided it by 100. I then made some graphs just to check the information I had collected was correct. Graphs are also easier to compare sets of data as the information is easier to read and stands out more.
From this chart I can see that actually the word length in a Tabloid is the longest. I didn't expect this, but after double-checking my working out, it appeared to be true. Also from this chart we can see that a Broadsheet uses the most 3-letter words, which again was unexpected.
A way of comparing information is by using the mean, median, mode or range. When we refer to the mean of something we are referring to the answer we get after adding up all of the data and dividing it by the amount of data we are using, for example, 2,3,6,4,8,12. The mean of this information would be: 2+3+6+4+8+12=35. 35/6=5.83r. When we refer to the mode of a number we a re really referring to the most frequent number in a list or set of data, for example, 2,5,7,4,2,5,9,2,1,2,5. The mode of this information would be 2 because it is the most frequent. When we refer to the median of a set of information we are really referring to the middle number of some information when in order of size, for example, 2,5,4,7,9,12,45,6,3,8 would be rearranged to form 2,3,4,5,6,7,8,9,12,45. Now we can find the middle number, which in this case is between 9 and 12. Now it's a case of finding the number that is between 9 and 12, which is 10.5. This is now the median. If there is a set of data with an odd number of information such as this 2,3,4,5,6,7,8,9,12,45,98 it is easy to find the median, as we just have to work our way to the middle, the median of this set of data is 7. Finally, when we refer to the range of a set of information we are really referring to difference between the largest and the smallest numbers, for instance, 2,1,3,4,6,4,7,5,6,9,23,12. The smallest number here is 1, and the largest is 23. If we do 23-1 we find the answer is 22. Therefore the range is 22.
When we compare the mean of the Tabloid with the mean of the Broadsheet we find that there is a difference of 0.82. This means that there is quite a substantial difference between the both; this also means that the numbers in the Tabloid must be bigger than in the Broadsheet. When we compare the median of both sets of data we get no difference in results, which means that the middle number of frequencies in both the Tabloid and the Broadsheet must be the same, or close to. When comparing the Mode of both sets of data we find that the mode is bigger in the Tabloid, it is quite a bit bigger which must mean there must be a bigger most frequent number in the Tabloid. When we compare the range of both sets of data we find that the range is quite a lot bigger in a Tabloid than in a Broadsheet, this tells me that there must be a bigger range of numbers in the Tabloid also it must mean that somehow either the smallest of the largest number must be bigger in the Tabloid in order for this theory to be true.
Sentence Length.
To obtain a sample for my sentence length I intend on using the same method of sampling as I used for the word length, this of course was systematic sampling. This means I will count number of sentences in an article and divide it by 100, this will tell me the nth term, which I can use to collect my sample. I have chosen not to use 'random sampling' and 'stratified sampling' as they would take to long. To make this a fair test I will ensure that the sentences I use are taken from newspapers on the same day about the same event. I have chosen to include sentences with connectives and punctuation such as commas (,) and semi colons (;), I have also chosen to include numbers, names and places in my research.
Focusing on sentence length, this is what I found for my tabloid and broadsheet.
Key: 1|2|6 = 21 and 26
Tabloid Broadsheet
I recorded my findings in a stem and leaf diagram because this way it is easier to read. From this diagram I can clearly see that the Broadsheet used longer sentences than the Tabloid.
I drew a box plot for each set of data to be 100% sure of my conclusions.
<- Broadsheet.
<- Tabloid.
When comparing both of the sets of information I can easily tell that the Broadsheet contains longer sentence lengths. I can tell this because the broadsheet box plot is further along the scale, which means there are longer sentences in it. I decided to draw some histograms to further compare the data.
Tabloid.
Broadsheet.
The histograms show that the broadsheet contains longer words than the tabloid also. Therefore I can confirm my earlier suspicions.
In conclusion my second hypothesis was correct whereas my first was wrong, during the first hypothesis my calculations do not support my hypothesis but in the second they do. If I was to repeat this experiment I may change the sample size or articles, maybe even the newspapers involved. Overall my results remain unaffected by any of the processes I have taken, the only time they would be affected would be if I had chosen to use names of places or numbers etc.
By Coco Smith