Write a hypothesis about the length of words in newspapers and magazines.
.0 Brief
Suresh is comparing magazines and newspapers.
He chooses a passage from one newspaper and one magazine. They each contain 100 words and he counts the lengths of all the words.
Suresh then says that the magazine has the shortest words.
* Write a hypothesis about the length of words in newspapers and magazines.
* Design and carry out an investigation to test your hypothesis.
Investigate Further.
2.0 Introduction
It has been stated from Suresh's previous study of comparing magazines and newspapers, that magazines have the shortest written words. In order to test this theory, two specific publications have been selected in order to prove this statement as false. In order for this to occur, It would be my expected opinion that a so-called 'higher standard' of publication would produce longer words in an article in comparison to 'lower standard' publication.
To test this theory, the two publications that have been chosen are the Hello magazine and The News of the World.
The 'Hello' magazine is currently the International magazine of the year. It has become highly popular on the coffee tables of hairdressers or with the aspiring housewife. Features include high society parties and exclusive celebrity wedding coverage. Hello is seen as a luxury publication that is sold on a weekly basis.
In contrast, 'The News of the World' is a high selling Sunday tabloid newspaper. The publication has become extremely popular with undercover reporting on celebrity scandal and kiss and tell stories. The paper would be mainly aimed at the middle - lower classes with manual or semi-manual professions.
Different publications aim at different target markets and are written to suit these preferences. We could suggest that both the selected magazine and tabloid are a light read in comparison to more descriptive broadsheet papers and specialist magazines. However, I believe that tabloid papers wording would be less profound and therefore more easily understood in relation to a high quality magazine.
This study will investigate the opening statement by analysing and concluding the results found from the data collected, whilst testing the stated hypothesis by using the same basic methodology that Suresh used.
3.0 Hypothesis & Objectives
A recognised 'High Quality' Magazine will have longer words than a recognised 'Low Quality' Sunday tabloid newspaper. In order to test the significance of this statement, research objectives need to be defined as outlined below.
3.1 Hypothesis
The quality magazine will have longer words than the tabloid newspaper.
3.2 Objectives
i. To collect data on number of letter per word in two publications.
ii. To present data in a meaningful way.
iii. To interpret and analyse results and diagrams.
iv. To draw conclusions on analysis, and state whether the hypothesis is correct.
The hypothesis and objectives will be the focus of the investigation following on from the study's methodology and data collection.
4.0 Methodology
The publications of Hello and The News of the World (NOTW) are both published on a weekly basis and the data was collected from the publications of January 7th 2003 and January 5th 2003 respectively.
In order to make it a fair test I randomly chose both articles out of each publication by closing my eyes and randomly selecting a page. Following this, the first article on this randomly selected page was chosen for the raw data collection.
The first 100 words from each publication were then counted to record the number of letters in each word for the collected data. The statistical analysis will follow three stages of sorting and grouping, illustration and summary statistics. The findings are presented firstly in table format to show the obvious differences between the two publications. Further to this, the results will be analysed in the form of the following:
> Pie Chart
> Bar Chart
> Frequency Polygon
> Cumulative Frequency Curve
> Overall Media, Mean & Mode
The purpose of any statistical analysis is to simplify large amounts of data, find any key facts and present the information in an interesting and easily understandable way. This investigation will produce such information in answering the study hypothesis.
5.0 Results and Analysis
The collection of the raw data for Hello and NOTW has been formatted into tables as shown in Appendix 1.1 and 1.2 respectively1. As shown, the words have been inputted into the tables so that the words would be represented in numerical order, the stated word, and the word length. The table was produced so that the information was easily and readily available to refer to throughout the investigation.
The data was then grouped by word length for each publication as shown in Appendix 1.3 and 1.4. After grouping the data into individual categories, the raw data has been represented in Table 1.5 for both publications, simplifying the total collection of data of 200 words that will be analysed in the investigation.
Table 1.5 Shows the tally for both publications.
Number of letters
in a word
Hello
The News of the World
Total
% Total
letter
4
5
9
4.5
2 letters
7
...
This is a preview of the whole essay
The data was then grouped by word length for each publication as shown in Appendix 1.3 and 1.4. After grouping the data into individual categories, the raw data has been represented in Table 1.5 for both publications, simplifying the total collection of data of 200 words that will be analysed in the investigation.
Table 1.5 Shows the tally for both publications.
Number of letters
in a word
Hello
The News of the World
Total
% Total
letter
4
5
9
4.5
2 letters
7
5
32
6
3 letters
7
23
40
20
4 letters
4
6
30
5
5 letters
7
9
26
3
6 letters
3
7
20
0
7 letters
7
4
1
5.5
8 letters
5
8
3
6.5
9 letters
4
9
3
6.5
0 letters
3
4
2
3 letters
2
Total
00
00
200
00
From table 1.5, the first piece of data that has been interpreted is that both publications share the same range of letters per word (LPW). The range is between 1-12, however both publications do not provide words with 11 or 12 letters.
The second observation from this table is that the highest percentage of letters in a word for the publications combined is 3 with 20 per cent. This was followed with 2 LPW with 16 per cent.
5.1 Results and Analysis
From the 'Hello' magazine data collection that is represented in table 1.5, we can identify that the LPW range is between 1-13 therefore the actual range is 12 letters. We can also state that the highest frequency value of LPW occurs for words with 2, 3, and 5 letters, with a frequency of 17 occurrences. The most commonly used diagrams are the various forms of bar charts. Extra
5.2 Bar Chart Analysis
The bar chart shown above allows us to identify more easily the spread of the data collected for Hello magazine. We can clearly see that the majority of LPW is focused between 2-6 LPW, with the higher number LPW tailing off between 7-13. With the equal highest frequency of 17, we can recognise a continual decrease from 5 LPW through to 13 LPW. The lowest recorded frequency is 1 occurrence, and is found for the highest values of 10 and 13 LPW.
It could be suggested from studying the bar chart of 'Hello' magazine, that lower frequency LPW are more commonly used.
In comparison, we can identify numerous differences from the News of the World newspaper bar chart shown below in Graph 1.2.
Firstly, the highest frequency value of LPW occurs for 3 letters with 23 occurrences. This frequency is significantly higher than the highest value recorded for Hello magazine. However, the highest value for both publications have been recorded at 3 LPW. In connection with Hello magazine, the lowest value is recorded at 13 LPW with a frequency of 1.
The NOTW graph with the linear trend line shows a much flatter decreasing line than that of Hello magazine. From this trend line and the graph, we can identify that the spread of LPW is greater for the NOTW in comparison to 'Hello' over the range of data collected, especially towards the higher frequency LPW.
However, we can state that a relationship occurs between the three highest frequency values for the NOTW of 2, 3, and 4 LPW being the most common as these are the joint first and fourth highest recorded for Hello's data.
From the bar chart, we can suggest that the News of the World use a wider spread of letters per word than Hello magazine for the range of data collected.
5.3 Pie Chart Analysis
From analysing pie chart 1.1, we can show how a total amount is divided into constituent parts. We can therefore back up our previous statement from the bar chart by suggesting that LPW 2, 3, 4, 5 and 6 represent the highest proportional areas with over three quarters of the total value. From the chart we can also see that other LPW have similar proportions. For example, 1 and 9 LPW have the same frequency of 4, whilst 8 LPW having a frequency of 5. Similarly, words with 10 and 13 letters only have a frequency of 1.
From looking at the News of the World pie chart (Pie chart 1.2 - Page ##) we can identify that once again, the proportion of the data collected is represented between 2-4 LPW, nevertheless the LPW from 5 -13 are more evenly spread in comparison to the pie chart 1.1.
A similarity that has been observed from the both pie charts is the proportion that is held between the constituent parts of 1 - 5 LPW, which appears to be identical. From further investigation of the tally, we can note that Hello magazine holds a 69 per cent proportion of its data between 1 - 5 LPW whereas, the NOTW holds a 68 per cent proportion between the corresponding LPW.
5.4 Data Analysis
Table 1.6 Represents the Cumulative Frequency for Hello magazine
Number Of Letters
Frequency
Cumulative Frequency
Cumulative Frequency Total
4
4
4
2
7
4+17
21
3
7
4+17+17
38
4(*)
4
4+17+17+14
52
5
7
4+17+17+14+17
69
6
3
4+17+17+14+1+13
82
7
7
4+17+17+14+1+13+7
89
8
5
4+17+17+14+1+13+7+5
94
9
4
4+17+17+14+1+13+7+5+4
98
0
4+17+17+14+1+13+7+5+4+1
99
3
4+17+17+14+1+13+7+5+4+1+1
00
(*) - The Median can be found within this section
Table 1.7 Represents the Cumulative Frequency for the NOTW newspaper
Number Of Letters
Frequency
Cumulative Frequency
Cumulative
Frequency Total
5
5
5
2
5
5+15
20
3
23
5+15+23
43
4(*)
6
5+15+23+16
59
5
9
5+15+23+16+9
68
6
7
5+15+23+16+9+7
75
7
4
5+15+23+16+9+7+4
79
8
8
5+15+23+16+9+7+4+8
87
9
9
5+15+23+16+9+7+4+8+9
96
0
3
5+15+23+16+9+7+4+8+9+3
99
3
5+15+23+16+9+7+4+8+9+3+1
00
(*) - The Median can be found within this section
The mean measurement takes into account every item of data collected and produces the average.
The mean number of letters in a word for 'Hello' magazine is calculated;
(1x4) + (2x7) + (3x17) + (4x14) + (5x17) + (6x13) + (7x7) + (8x5) + (9x5) + (10x1) + (13x1)
00
= 4.56 mean
The mean number of letters in a word for the NOWT is calculated;
(1x5) + (2x15) + (3x23) + (4x16) + (5x9) + (6x7) + (7x4) + (8x8) + (9x9) + (10x3) + (13x1)
00
= 4.71 mean
We can state that on average 'Hello' magazine uses 4.56 LPW whilst the mean average for the NOTW is 4.71. Therefore, NOTW averages a slightly higher frequency of letters per word.
The modal number is the value that occurs most frequently. As previously discussed, the modal numbers are 2, 3 and 5 with 17 LPW. In comparison, the modal number for the NOTW is 3 with 23 LPW.
This data represents that the modal number for the NOTW is higher, although 'Hello' magazine has three modal numbers in comparison which includes the same modal number as NOTW plus a higher LPW modal number of 5.
The median is the middle value of the set of data. From the data set of 100 the median is represented by the value of 50. From table 1.6 and 1.7 we can see that the median is represented by 4 LPW for both publications. From this data we can not state that either publications have a higher or lower median value.
5.# Frequency Polygon and Cumulative Frequency Analysis
From grouping the collected data, we can assume an even spread of data within each group in comparison to the previous analysis. In order to analyse the grouped data, we have presented this in the table shown below and then represented each as a frequency polygon and cumulative frequency diagram.
Table 1.8 Represents the Grouped data and cumulative frequencies for both publications.
Grouped -
Number of letters
in a word
Frequency -
Hello
Cumulative
Frequency
- Hello magazine
Frequency -
NOTW
Cumulative
Frequency
- NOTW magazine
to 3
38
38+0 = 38
43
43+0 = 43
4 to 6
44
44+38 = 82
32
32+43 = 75
7 to 9
6
6+44+38 = 98
21
21+32+43 = 96
0 to 12
+16+44+38 = 99
3
3+21+32+43 = 99
3 to 15
+1+16+44+38 = 100
+3+21+32+43 = 100
From the two frequency polygons for Hello and NOTW represented in Appendix ## and ## respectively, we can see that a similar shaped frequency polygon. This is mainly due to the fact that the last two groupings are identical of 10 to 12 and 13 to 15, with a cumulative frequency of 99 and 100 LPW respectively. Plus the third grouping of 7 to 9 is very similar with a cumulative frequency of 98 for Hello and 96 for the NOTW.
However, we can note that Hello's frequency polygon between the first three groupings is far steeper than that of the more gradual increase of the NOTW's frequency polygon. Therefore as previously stated, the spread of data is more evenly distributed for the NOTW's grouped data.
From the groupings we can also state that Hello magazine has the majority of higher grouped LPW in comparison to the NOTW.
The grouped data also allows us to analyse the results in the form of a cumulative frequency graph shown in appendix ## and ## for Hello and the NOTW respectively. The cumulative curve is found by plotting the cumulative frequencies and joining them with a smooth curve.
Once again we can state that the curve for the NOTW is a little bit flatter than the steeper curve of Hello. Looking at the graphs more closely we can identify the upper, inter and lower quartile ranges.
The lower quartile range is found at approximately 2.1 LPW for Hello, whilst the NOTW's is approximately 1.6 LPW. The inter quartile range is found at approximately 3.6 LPW for both publications whereas the higher quartile range is greatest the NOTW. This is found to be approximately 5.9 LPW whilst Hello's is approximately 5.4 LPW.
The findings from the cumulative frequency graphs show us that Hello magazine has a greater proportion of lower LPW whilst the NOTW has a greater proportion of higher LPW.
6.0 Conclusions and Recommendations
At the beginning of the investigation, it was believed that magazines have the shortest words for the study conducted by Suresh. I believe this statement to be wrong and in order to test this hypothesis incorrect, two publications were selected at either end of the market to prove the new hypothesis: The quality magazine will have longer words than the tabloid newspaper.
However, from the analysis of the graphs and manipulation of the raw data it was found that my hypothesis was to be incorrect. The tabloid newspaper was found to have longer words in the selected article than the upper market magazine. Therefore this agrees with the original hypothesis stated by Suresh.
In order to test my hypothesis further, ideally it would be more conclusive to analyse numerous articles from the same publications. This would allow a wider collection of data, and then improve the analysis and draw more accurate conclusions.
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. Although, I still believe that this could be proved correct with further investigation. Alternatively, this extension will focus on, the number of words per sentence (WPS) in contrast to letters per word.
7.1 Hypothesis
The quality magazine will have more words in a sentence than the tabloid newspaper.
The objectives and methodology used in the original investigation will be applied to this extension.
7.2 Results and Analysis
The collection of the raw data for Hello and NOTW has been reformatted into tables as shown in Appendix ## and ## respectively. As shown, the words have been inputted into the table so that the words would be represented in numerical order, the stated word and the number of WPS. The table was produces so that the information was easily and readily available to refer throughout the investigation.
The Hello article consisted of only 13 sentences, therefore to make the investigation a fair test, the first 13 sentences of the NOTW were used.
Table 2.1 Shows the tally for both publications
Number of words in a sentence
Hello
The News of the World
Total
st sentence
23
22
45
2nd sentence
26
29
55
3rd sentence
38
25
63
4th sentence
31
7
48
5th sentence
9
23
42
6th sentence
32
2
44
7th sentence
35
7
42
8th sentence
32
8
40
9th sentence
7
8
25
0th sentence
7
20
37
1th sentence
28
33
51
2th sentence
24
25
49
3th sentence
27
41
58
Total
349
270
619
From table 2.1, the first piece of data that appears is that the highest number of WPS for both publications occurs in the 3rd sentence. This coincides with the highest for Hello with 38 WPS. In contrast, the NOTW's 3rd sentence is the fourth highest WPS. The NOTW's highest recorded at 41 WPS.
2.1 Bar Chart Analysis
The bar chart above allows us to identify the highest number of WPS for 'Hello', which appear in the 3rd, 4th, 6th, 7th, 8th and 12th sentences. The bar chart does not show a clear spread of data as the frequency increases & decreases quite regularly throughout the graph. We can recognise a similar pattern throughout the graph, starting from the first sentence. Each two sentences after the 1st sentence increase in frequency then as the 3rd sentence is reached the frequency decreased again till it reached the 6th sentence, this same pattern is repeated until we reach the 13th sentence. This tells us that there is a continuous increase followed by decrease throughout the data. The equal lowest frequency 17 occurs, both in the 9th and 10th sentence. The highest recorded frequency is 38 and is found in the 3rd sentence.
In comparison, we can identify numerous differences from the NOTW tabloid newspaper bar chart shown below in Graph 2.2
From looking at the bar chart it is noticed that the highest frequency value for WPS is 58 which appears in the 13th sentence. This is significantly higher than the highest value recorded for 'Hello' magazine of 38 WPS. And both frequencies are found in different sentences NOTW being found as mentioned before in the 13th sentence and 'Hello' being found in the 3rd sentence.
From looking at the two graphs we can see that the NOTW uses less number of WPS than 'Hello'. As the frequency is obviously higher throughout the 'Hello' bar chart.
?.? Data Analysis
Table ?.? Represents the Cumulative frequency of 'Hello' magazine
Number Of words
Frequency
Cumulative Frequency
Cumulative Frequency Total
23
23
23
2
26
23+26
49
3
38
23+26+38
87
4
31
23+26+38+31
18
5
9
23+26+38+31+19
37
6
32
23+26+38+31+19+32
69
7
35
23+26+38+31+19+32+35
204
8
32
23+26+38+31+19+32+35+32
236
9
7
23+26+38+31+19+32+35+32+17
253
0
7
23+26+38+31+19+32+35+32+17+17
270
1
28
23+26+38+31+19+32+35+32+17+17+28
298
2
24
23+26+38+31+19+32+35+32+17+17+28+24
322
3
27
23+26+38+31+19+32+35+32+17+17+28+24+27
349
?.? Pie Chart Analysis
From analysing pie chart ?.?, it is noticed that the proportions have very similar frequencies throughout. All numbers of WPS have very little difference in frequency between them ranging from 1 to 3. For example the difference between the first sentence
Table ?.? Represents the Cumulative Frequency for the NOTW newspaper
Number Of words
Frequency
Cumulative Frequency
Cumulative Frequency Total
22
22
22
2
29
22+29
51
3
25
22+29+25
76
4
7
22+29+25+17
93
5
23
22+29+25+17+23
16
6
2
22+29+25+17+23+12
28
7
7
22+29+25+17+23+12+7
35
8
8
22+29+25+17+23+12+7+8
43
9
8
22+29+25+17+23+12+7+8+8
51
0
20
22+29+25+17+23+12+7+8+8+20
71
1
33
22+29+25+17+23+12+7+8+8+20+33
204
2
25
22+29+25+17+23+12+7+8+8+20+33+25
229
3
41
22+29+25+17+23+12+7+8+8+20+33+25+41
270
The mean measurement takes into account every item of data collected and produces the average.
The mean number of letters in a word for 'Hello' magazine is calculated;
(1x23) + (2x26) + (3x38) + (4x31) + (5x19) + (6x32) + (7x35) + (8x32) + (9x17) + (10x17) + (11x28) + (12x24) + (13x27)
3
= 182.4 mean
The mean number of letters in a word for the 'NOTW' tabloid newspaper is calculated;
(1x23) + (2x26) + (3x38) + (4x31) + (5x19) + (6x32) + (7x35) + (8x32) + (9x17) + (10x17) + (11x28) + (12x24) + (13x27)
3
= 153.2 mean
We can state that on average 'Hello' magazine uses 182.4 WPS whilst the mean average for the NOTW is 153.2. Therefore, the NOTW has a lower frequency of words per sentence.
The modal number is the value that occurs most frequently. As previously discussed, the modal numbers are 6 and 8 with 32 WPS. In comparison, the modal number for the NOTW are 3 and 12 with 25 WPS and also 8 and 9 with 8 WPS.
This data represents
The median is the middle value of the set of data. From the data set of 349 the median is represented by the value of 174.5. From the table ?.? and ?.? we can see that the median is represented by 7 WPS for both publications. In comparison the data set from the NOTW is 270 and is represented by the value of 135.
?.? Conclusions and Recommendations
Two hyphenated words were found in both articles of each publication and have been counted as two separate words.
Read All About It by - Joanne Morgan
- 1 -
