*From these tables I can see that the broadsheet has a longer word length to the tabloid. This is not very easy to see however and because of this I will display this information in a clearer form
*From these tables I will now create a cumulative frequency table for both newspapers
Cumulative frequency table:
Broadsheet:
MEDIAN: frequency +1 = 50+1 = 25.5
2 2
LOWER QUARTILE: frequency+1 = 50+1 = 13
4 4
UPPER QUARTILE: lower quartile x 3 = 13x3 = 39
Tabloid:
MEDIAN: frequency +1 = 50+1 = 25.5
2 2
LOWER QUARTILE: frequency+1 = 50+1 = 13
4 4
UPPER QUARTILE: lower quartile x 3 = 13x3 = 39
*From this cumulative frequency table, I am now going to create a cumulative frequency graph, which will include the median, and upper and lower quartiles.
*From my cumulative frequency graph (on graph paper) I can see that the mean is higher for the broadsheet than the tabloid. This agrees with my hypothesis.
*I believe that this information would be easier to read if It was in a box and whisker diagram. Therefore I am going to create one displaying this information.
*I believe this box and whisker diagram (graph paper) shows very clearly that the broadsheet has a higher median and higher upper and lower quartiles.
Relative frequency:
The relative frequency is the probability of getting a long or short word from my random samples. Having looked at my results, I have decided that a long word is a word with more than 10 letters whereas a short word is a word made up of 5 or less letters. By using these margins and displaying information in a table I am going to work out the percentage chance of picking a long or short word at random
To find the percentage chance of picking a long or short word at random I will divide the frequency by the total sample size, then times this figure by 100. When I have done this for all the values I will display my results in a table.
*This table agrees with my hypothesis, the average word length is longer in a broadsheet. At the same time, the tabloid has a higher percentage of short words than the broadsheet. This backs up my hypothesis even more.
*I had intended to go on to make a stem and leaf diagram but having looked at my results, have realised that this is not appropriate. This is because it will not tell me anything I haven’t already discovered, the other diagrams and tables are enough to tell me that the broadsheet has a longer average word length, this has successfully proven my hypotheses. Also I don’t believe that the scope is large enough to justify making a stem and leaf diagram. With all this in mind I have decided not to do a stem and leaf diagram.
Evaluation and conclusion:
Having done this work I have been able to successfully prove my hypothesis. All the diagrams I have made agree with my hypothesis.
The cumulative frequency table and graph completely back up my prediction, that the broadsheet has a longer average word length. From looking at my graph I can clearly see this but I decided that it would be easier to read if it was displayed in a box and whisker diagram. By looking at the both of these diagrams I can clearly see that the broadsheet has a higher median and a higher lower and upper quartile. I am pleased with this because it helps to prove my hypothesis and my prediction
From my cumulative frequency graph (on graph paper) I can see that the mean is higher for the broadsheet than the tabloid. This agrees with my hypothesis.
I believe this box and whisker diagram (graph paper) shows very clearly that the broadsheet has a higher median and higher upper and lower quartiles.
From looking at both of these diagrams I can tell that the broadsheet has a longer average word length but I still didn’t know what the percentage chance of picking a long or short word was. I could see that the broadsheet had a longer average word length from looking at my cumulative frequency and box and whisker diagrams, but I wanted to prove that the percentage chance of picking a long word was higher in a broadsheet. To do this I decided to make a relative frequency table. I am pleased with the way my relative frequency table has worked out because it proved what I had thought, that the percentage chance of picking along word was higher in a broadsheet than in a tabloid.
Not only this, but it also proved that the percentage chance of picking a short word was higher in a tabloid than in a broadsheet. All of this information helps to back up my hypothesis and prediction, that the average word length is higher in a broadsheet than in a tabloid.
Overall I, by looking at all my work, I can successfully prove that the average word length is higher in a broadsheet than in a tabloid.
To further my knowledge of broadsheets and tabloids, I have decided to do a second hypothesis. The second hypothesis will be related to the first and will be about sentence length instead of word length.
Hypothesis 2:
The average sentence length is higher in a broadsheet than in a tabloid.
Plan:
First I will number the articles in each newspaper, I will ensure I number 50 articles.
Using the random button on my calculator to avoid bias, I will look at a particular sentence in the article. For example, if the calculator displays the number, 0.374 I will look at the third sentence in the article and count the amount of words in this sentence. To avoid bias I will always look at the first number after the decimal point. However if the first number after the decimal point is a zero I will repeat the method until I get a number from 1-9.
I will repeat this process for every numbered article in the newspaper and then record my results, I will do this for both the broadsheet and the tabloid.
Having done all this I will compile more graphs and tables, much like I did for the first hypothesis. However I am not going to go into such detail in my second hypothesis, I am only going to create a cumulative frequency graph and a box and whisker diagram. I am hoping this will be enough to prove my second hypothesis. Of course, if this isn’t enough I will go into more detail, perhaps making a stem and leaf diagram and possibly a relative frequency table.
Collecting the number of words in every sentence of the newspaper would be too difficult and time consuming.
Therefore I am going to need to sample my sentences. To do this I am going to use the random button on my calculator. I will then be able to take a random sentence and record my results fairly, avoiding bias. For my first hypothesis I used a sample size of 50 however for my second hypothesis I am going to use a sample size of 30.
I predict that the broadsheet will have a longer average sentence length than the tabloid.
Data collecting:
Broadsheet:
Tabloid:
*from these tables I can see that the broadsheet has a longer average sentence length but I find this hard to read.
*using the information displayed in these tables I am now going to make a cumulative frequency table
Cumulative frequency table:
broadsheet:
MEDIAN: frequency +1 = 30+1 = 15.5
2 2
LOWER QUARTILE: frequency+1 = 30+1 = 7.75
4 4
UPPER QUARTILE: lower quartile x 3 = 7.75 x 3 = 23.25
Tabloid:
MEDIAN: frequency +1 = 30+1 = 15.5
2 2
LOWER QUARTILE: frequency+1 = 30+1 = 7.75
4 4
UPPER QUARTILE: lower quartile x 3 = 7.75 x 3 = 23.25
*Having made my cumulative frequency table I am now going to display the information in a cumulative frequency graph. I believe that the information displayed in the table would be easier to read if it was in a graph.
*My cumulative frequency graph (graph paper ) was a success and it shows information which agrees with my prediction and hypothesis
*However I believe it would be easier to read if it was displayed in a box and whisker diagram.
*The box and whisker diagram (graph paper) was also a success and I find it much easier to read than the cumulative frequency graph, it clearly shows that the broadsheet has a longer average sentence length and this agrees with my hypothesis.
*The difference between the two papers is even larger for my second hypothesis.
Evaluation and conclusion:
Having done all this work I have been able to prove my hypothesis. I have also been able to prove my prediction was right. I have been able to display lots of different information in many different diagrams and tables. All of which have helped me prove my hypothesis.
As I have done two hypotheses’ I have been able to come to a general conclusion. The first hypothesis proved that the average word length was larger in a broadhsheet. The second hypothesis proved that the average sentence length is larger in a broadsheet. By looking at both these results I have been able to see that a broadsheet newspaper has a more intelligent content than a tabloid.
If did this investigation again I think I would use a larger sample size, this would enable me to use a larger variety of methods to prove my hypothesis, for example, I couldn’t do a stem and leaf diagram because I felt the scope wasn’t large enough.