Friday, March 21, 2014

Misleading Statistics (Term 1, 2014 Monday Wednesday Friday 6:00 PM Class)


            Misleading statistics, which are false data based on studies, are used to confuse people (Cambridge advanced learner’s dictionary). The first statistic was published in 1663 (Wikipedia). Since then, statistics have been widely used by millions of people all around the world in many cases including business, education, advertisements et cetera. However, statistics can be misleading, so they should be interpreted with caution. This essay aims to demonstrate various strategies that can make statistics unreliable such as sampling bias, misleading graphs, misleading advertisements and unfair comparisons.



Australian Centre for Education
Writing Assignment
           
Level: EAP5                                                                   
Day: Mon/Wed/Fri
Time: 18:00-19:30

Sampling Bias
Response bias usually creates a misleading statistics. Due to its procedure, which is applied by interviewers to collect data from participants in the survey. The information is usually not trustworthy because of two main factors. Firstly, interviewers are questioning the people face to face. So the intention of participants to avoid humiliation, the participants are likely to provide false information. Secondly, completing a questionnaire publicly causes the information to have no confidentiality. Therefore, it is unreliable data as well (Sternstein, 2010). We cannot obtain the true information from the following survey such as the level of girls personal hygiene, a doctor are not revealing how confident he is to cure a disease, and a couple seemingly lie about the problems that they have in their married life.

Healthy user bias is also one among the other factors that could make a misleading statistic. It means that they randomly pick only healthy people to do the test and the result will be the data, which will be used to represent the whole thing. Normally, the first reason is when researchers want to do research on how healthy people are, the researchers aim to go somewhere crowded such as a company, an office, or a supermarket, and so forth. They will randomly select a number of people.   Normally, people, who are picked by the researchers there, are healthier than people who stay at home (William, H. S., Amanda, R. P & M, A. B, 2011),because sick people may stay at home to get a full rest, and take medication. Secondly, if they want to lie about the information on how easily people would be at risk of diseases, researchers would select only old people to the do survey because they are lack of vaccination (Sturmur, n.d). For example, if they want to exaggerate about the rate of cervical cancer, they will select only old women to do the survey. Third, in case of doctors aim to study about the disease’s symptom, while patients have been using many different drugs, so the result might be inaccurate because that the symptom might not appear in the period of medication. Furthermore, the result might be shown as a false negative, which means those that do have diseases and tests are shown as negative (Jo, Raquel et al., 2009) For instance, infectious diseases might not have a fever as they might have used the anti-fever drugs.
            Self-selection is another method that can lead to an invalid sample. It can be defined as a kind of sample that is uncertain and biased, and which is a serious problem leads to a misleading statistics due to the small sample size, voluntary sample, the accessibility of the participants and the area where the sample is obtained. Respondents of a survey have an option to choose whether they would like to take part in the survey or not. This can make an unequal and unacceptable result. In some cases, people with a strong opinion regarding the topics will be more likely to participate. The sample can be biased due to geographical and technological development differences. For example, a telephone sampling includes only people with cell phone, so the other people cannot participate in this kind of surveys, especially in remote areas where the majority of the population over there cannot afford to own a cell phone (Hudson, Seah, Hite &Tim, 2004). Another example is the Internet polls, where only the Internet users or people from the areas where Internet access is available can take part. As a result, theo btained sample can be inaccurate, and cannot represent the whole population needed in the study. Moreover, another problem within this sample is the attempt to manipulate the result of the survey by a particular group of people. An example can be seen in the voting of Book of the Year Award in the Netherlands in 2005, which the winner was the Bible translation due to a movement by the Christian society. Therefore, this group of people cannot represent the whole population of the Netherlands. (Bethlehem, 2008).

Exclusion bias is one of the problems that occur in statistics in which someone or something does not get included in a study or data. In this case, let’s take a look at some imaginary examples-a group of media students plan to do a survey in X area base on the amount of people in the area that enjoy the media such as watching films, listening to broadcast news or reading newspapers and magazines. However, when the research is done, they find out that five families are not included in the data because they just moved out of the X area, although their names remain listed in the registration. Therefore, the five families are stated as the subject of exclusion bias. Moreover, another error might occur when individual Z happens to be in the X area for several days, but haven’t gotten their names listed in the registration. In both moving-in and moving-out cases, they can be listed as exclusion bias.














Bibliography:
1.      Sternstein, M. (2010). Barron's AP Statistics. (p. 169). New York: Barron's Educational Series, Inc.

2.      William, H. S., Amanda, R. P & M, A. B. (2011). Healthy User and Related Biases in observational Studies of Preventive Interventions: A Primer for physicians. J Gen Intern Med, 26 (5), 546-550.
3.      Sturmer, Til (Unknown). Lesson about Confounding and Selection Bias Taught by the New User Design [PDF file]. Available from: http://www.pharmacoepi.org/pub/1c22d3d0-2354-d714-5186-0d8a16a5ffa6[Accessed 11February 2014].
4.      Jo, Raquel, et al., (2009). Clinical tests. A Systematic Reviews. Available from:https://www.york.ac.uk/inst/crd/SysRev/!SSL!/WebHelp/2_2_DIAGNOSTIC_TESTS.htm[Accessed 9February 2014].
5.      Hudson, D., Seah, L., Hite, D., &Tim, H. (2004). Applied economics letters. (4 ed., Vol. 11, pp. 237-240). Retrieved from http://www.tandfonline.com/doi/abs/10.1080/13504850410001674876[Accessed 9February 2014].
6.      Bethlehem, J. (2008). How accurate are self-selection web surveys? Retrieved from http://www.cbs.nl/NR/rdonlyres/EEC0E15B-76B0-4698-9B26-8FA04D2B3270/0/200814x10pub.pdf[Accessed 9February 2014].




Misleading Graph

            Statistics are used in almost every aspect of people’s everyday lives and play an important role of illustrating data. People tend to explain most complicated figures into graphs so that it is much easier to understand. However, the information can be exaggerated, and it misleads intentionally or accidently in many ways. Misleading graphs can deceive people and lead them into a misunderstanding perspective of the charts. Frequently, scientists, experts, or researchers, seem to use misleading graphs to prove their work in a significant way. According to a statistic expert, the writer of the book which relates to the introduction of statistics said that the disingenuous graph is constructed in an inaccurate portrayal or any confusing ways just in order to covey the wrong ideas to the users (Kirk, 2008, Internet).The aim of this essay is to demonstrate the misleading graphs in various ways such as scaling and axis manipulation, biased graph and 3-dimensional graph, by supporting with the example in order to provide more accurate ideas about deceptive displays. 
            First of all, the most common use of misleading graphs is scaling and axis manipulation. This sort of graph works along with vertical and horizontal axis; it usually has the same scale in each unit so that the graph looks unique and reliable (Utts, 2005, Internet). Nevertheless, the way of using this type of display may result in deceptive information and failure. For instance, if we talk about the stock market of a company that has a slight decrease or increase, the graph will show almost a steady line with no fluctuation. Thus, they may increase the proportion either on X or Y-axis intentionally in order to make it look like it has decreased or increased more in a significant way and attracts the viewers. This kind of graph is also called the gee-wiz graph (Haney, 2011, Internet).

Figure 1: Stock Price
Moreover, truncated graphs can also result in misleading information. Regarding the following graph, it can be noticed that the average house price does not start form “0” on Y-axis and the bar chart demonstrates the average house price is £80,000 and £82,000 in 1998 and 1999.Without paying much attention on it, we may assume that the price trebled within only one year, but in reality, it does not. It is just a slight change in the way of displaying the graph, and it can absolutely make people have a biased perspective about the information .Furthermore, if the truncated bar graph is similar to the old one, and a part of graph is cut so that the short bar graph will probably appear a little from x-axis (Rensbergers, 2009, Internet). That kind of graph can give information in wrong ways and misleading data, since it may show an extreme difference of the two graphs. In fact, it is the only way they were trying to show the same figure, but in the artificial way in order to make people have a new confusing idea about it.

Figure 2: House Price

            The second most used misleading graph is the biased graph. It includes the use of two or more different sizes pictures. Although there are two different sizes pictures with the same amount of data, people tend to put their eyes on the bigger size onerather than the ordinary ones. Some graphs are made in this technique to deceive viewers to believe that the result of sales, works, annual incomes, stock markets or even the recycle products increase dramatically over some periods of time.For example,the following picture shows that the trash is enlarged significantly over a short period. It means that the same pictures are magnified to make a higher height and a wider width. If we look at the area of the picture and compare the trash in 1960 and 1980, we may think that the trash increased 3 times while the real one was about two times only (Haney, 2011, Internet).This type of chart is one of the most disingenuous displays that people usually do, since it will probably enable the maker to convince the viewers to believe in the inaccurate information. 
           
Figure 3: Trash Amount

            The third of misleading graph is 3-dimensional graph, which is the most attractive and beautiful designed graph, as it gives people the ability to see the picture in 360 degrees (Rumsey, 2010, Internet). In contrast, advantages and disadvantages always come together.The more authentic the graph is, the more misleading it becomes. From the following picture, it can be considered that there is no scale on vertical axis and the sales in 1995 are much higher than the sales in any other years. In fact, it is identical to 1997 if we see it in 2-D graph.


Figure 4: Number of sales from 1995-1998


Last but not least, in 3-D pie chart below shows that the most numerous pie is D followed by item B, then item C and the least is item A. In contrast, this information is totally wrong from what we can see in the real regular pie chart. Item D is 42% which is the same as item B. However, the 3-D pie expands the front picture so that it looks bigger and the back one becomes smaller.

Figure 5: 3D and 2D pie charts
To sum up, all of these three types of misleading graphs are what people should be careful of, as we may be tricked bythem every time in everything that we are living with.  Misleading graphsdo not only trick its viewers, but also spins their heads around to misunderstand the real things.
























Biography

1.      Haney, B.R., (2011). ‘How to Lie With Statistics.’Math 143 Project[Online]. (Springed),Calvin College.Chapter 5, 6. Available from:http://www.calvin.edu/academic/economics/faculty/bios/HaneyDocs/ [Accessed 1 February 2014].
2.      Kirk, R.E., (2008), ‘Frequency Distributions andGraphs’ [online]. Statistics: An Introduction. (5thed). Belmont, USA, Thomson Higher Education. Pages 52-54. Available from: http://books.google.com.kh/books?id=W4t9Nfgk03AC&printsec=copyright&source=gbs_pub_info_r#v=onepage&q&f=false [Accessed 4 February 2014].
3.      Rensberger. B., (2009).  ‘Slanting the Slope of the Graph.’ The Washington Post [Online] 10 May. Available from: http://www.highbeam.com/doc/1P2-831228.html [Accessed 3 February 2014].
4.      Rumsey, D.J., (2010). ‘Ten Common Statistical Mistakes.’ Statistics Essentials for Dummies [Online]. Indiana, USA, Wiley Publishing Inc. Pages 155-162. Available from: http://books.google.com.kh/books?id=QBmsVY0p7YkC&pg=PA155&redir_esc=y#v=onepage&q&f=false [Accessed 2 February 2014].  .
5.      Utts, J.M., (2005). Seeing through statistics [Online]. (3rded). Belmont, USA, Thomson Brooks/Cole. Pages 146-147. Available from: http://books.google.com.kh/books?id=j5xWsf4DD58C&printsec=frontcover&source=gbs_vpt_buy#v=onepage&q&f=false [Accessed 28 January 2014].





Word Count: 871

Level: EAP 5                  
Time: 18:00-19:30
Room: G.3



FINAL DRAFT

Advertisements are usually win-win methods for both companies to make their products visible for people around the world and customers to find what they are in need of. However, there are some situations when they become win-lose methods, supposedly when they are misleading. There are many methods to make an advertisement misleading. This essay will thoroughly describe how different methods works on each situation and give an example for each of them.

The first method is called “Guarantee without a Remedy Specified”. The technique is used when companies do not specify what they will make up to the customer, if the products are under expectation. When the product fails, unlike what the customers expected, the companies are free to do very little. However, the law says that the advertisers have to be clear on advertising, not only to determine an error but also to give a solution for the error (“False advertising,” 2014). For instance, the chill tonic advertisement (see Figure 1) below does not give any specific remedy because it only mentions, “No cure No pay”.



Fig. 1 A guaranteed-without-a-remedy-specified advertisement

The second method to make misleading advertisements is “No risk”. It is a strategy when the advertisers claim that there are no risks to try their products when clearly there are. For example, they may charge the customers’ credit cards for the products and they will offer a full refund if not satisfied. However, the risk of this kind of offer is quite enormous. The customers may not receive the products at all or the products can be something that the customers did not paid for. Sometime, they tell the company to permit a return but they are unable to do so (“False advertising,” 2014).

The third method of misleading advertisements is “Hidden Fees”. It is a confusion caused by products’ owners, who try to hide extract fees in small letters or use confusing terms (“False advertising,” 2014). It is an attractive method to advertise in order to gain the amount of the product sold. Obviously, it can lead to the misleading statistic when the prices that are shown on the products are not fully covered all of the amount of expense the customers will make, which means they may have to spend more for other services, such as taxes and step-by-step payments. As we can see from the picture (see Figure 2), which is an advertisement of a smart phone called iPhone. It shows the lowest price of the original price on the advertisement. However, this is a step-by-step payment method. This can confuse and attract consumers. Moreover, that price may exclude the taxes payment. This method can decrease the price shown on the advertisement.


Fig. 2 An iPhone advertisement

     The next principle method is “Bait-and-switch”. It is used when advertisers advertise an unavailable item to attract the clients to visit theirs shops. When the consumers arrive at the stores, they will be convinced to buyother similar products at higher price. “Bait-and-switch” is legal in some countries, especially in the United States (“False advertising,” 2014). For example, the products that are advertised on newspapers at special occasion such as New Year or Christmas sometime are not available at the stores. However, by the time the customers learn about the unavailability, it will be too late since they will be at the store already. So they will be forced to try other products.

     Similarly, there is another method to mislead advertisements. It is the “misuse of the word ‘free’”. “The usual meaning of ‘free’ is ‘devoid of cost or obligation’” (“False advertising,” 2014). Sellers use this word to give away products that the prices are already included in overall. The "buy one, get one free" sale is the most common example. The meaning of "free" of the second item is not normal, since, to get it, the consumers have to fully pay the fee of both items on the first item (“False advertising,” 2014). For instance, hotels advertisements such as “Stay two nights, get the third free” can be misleading because the price of the third night may be included in the first two night already.

Last but not least, comparative statistics can also be misleading advertisements. This method can be notified when advertisers use comparative words such as “better” or “more” in the advertisements. In this method, advertisers prove their products to be better but they usually left out what the products are better than.  When people read the advertisements with only half of the information was given, their brains will automatically create another half("Chapter 7:the semiattached figure,").  For example, this method was applied on a toothpaste brand called “Arm & Hammer” (see Figure 3). Clearly on the package, there were letter saying “3 shades whiter, clinically proven”. Normally, customers will think that the toothpaste will make their current teeth to become 3 shades whiter. However, this toothpaste might mean that it makes the teeth 3 shades whiter than not brushing or than using other products.


Fig. 3 A toothpaste package

All in all, advertisements are not trusted sources for people to believe and they are very dangerous for ignorance because they might get confused or misled by the advertisement. I believe that if there are no improvements in any time soon, the world will face an economic problem because people will stop believing in the advertisements and stop buying products. So further rules must be made to stop this type of lie and some other rules must be strengthen before it becomes a serious problem.






Bibliography

1.      False advertising. (n.d.). In Wikipedia. Retrieved February 25, 2014, from:http://en.wikipedia.org/wiki/False_advertising
5.      Chapter 7:the semiattached figure. (n.d.). Retrieved from: http://www.calvin.edu/academic/economics/faculty/bios/HaneyDocs/page-58978274.html




Unfair Comparison (4th draft)
            There are many forms of statistics that people use every day. Unfair comparison statistics are statistics that compare two unequal items or concepts. One data is always better than the other which is why it is misleading. Generally, this type of statistics is popular in the business world. Most companies utilize this statistics to bring out a congenial image for their products as well as to make the product sounds better than it really is. This statistic is also used in numerous government departments and some kind of reports in order to make up surprising outcome or how better one item is, compared to the other item. This type of statistics served as a beneficial tool for most companies or organizations. However, it can also cause serious issues for many people and companies. For instance, it can lead one company into an atrocious image which can possibly make a company go bankrupt or lose their businesses because being compared to a bigger or better company will make one company look worse than it really is. It is an extremely bad influence to societies because most people think that every information that comes out as a survey or a statistic is reliable and trustworthy. This is why a lot of confusion was created and it harms people’s health and property. Unfair comparison is technically right but if people think about it carefully, we will figure out that it is actually so misleading that we cannot believe in it. This type of statistic is very tricky; it can easily convince people to purchase something or to believe in something.
            Many examples of unfair comparison can be seen in statistics of every sort. One of these examples, used by crop fertilizing company, includes the comparison of crop yields on selected years (Beri,2010). This method is used in order to showcase how much improvement their products can achieve. It can be done by simply selecting the crop yield of the year when there are occurring disasters including drought or inadequate rainfall and compare it to the year when there is no such issues. False information, which states the differences between the crop production before and after the used of the fertilizing product, can then be added to convince the audience and show how much of an impact their products can make. This comparison will definitely give the product a better picture (Beri, 2010). Another example of unfair comparison occurs when there is a comparison of complaints between companies including airline business (Lori Alden, 2005-7).
















As shown in the figure above, it appears to viewers that the airline companies that got the most complaints which in this case is United Airlines, is the worst airline company; while the least complaints of airlines including the Alaska Airlines, Southwest Airlines and Continental Airlines are the best airline companies. This statistic is confusing since it does not provide any related information or reason why United Airlines got the most complaints. This could be due to numerous reasons including the amount of passengers on the United Airlines.
            Even though, the unfair comparison or comparing apples and oranges are complicated and misleading, there are many ways to make the statistics more accurate. Firstly, the writers can compare best information with best information (Lori Alden, 2005-7). Secondly, if the writers do not want to compare best information with best information or they did not get enough details about the best data consequently, they can compare lowest information and lowest information, according to Lori Alden (2005-7). It is another strategy to make the statistics more reliable (Lori Alden, 2005-7). Also Lori (2005-7), illustrates that if the writers want to make a statistic for their business such as their products, they can compare their products with another companies’ products by comparing all to all. So, all the details are necessary because everyone can see and examine it. Thus, it is better to compare all to all because some people might not want to see only best data with best data or lowest data with another lowest data but they want to know all the information both good and bad consequently, it is a good way to compare objects or data by using this strategy (Lori Alden, 2005-7). There are many strategies of making the accurate statistics and all of them are usually used in the statistics to make it more reliable because the unfair comparison or comparing apples and oranges might make people get the wrong information.



Bibliography:
-          Beri, G. C. (2010). Business statistics. (3rd ed.). New Delhi: Tata McGraw Hill Retrieved from http://books.google.com.kh/books?id=tWmoP49v1cIC&printsec=frontcover&source=gbs_ge_summary_r&cad=0

-          Lori, A. (n.d.). Statistics can be misleading. Retrieved from http://www.econoclass.com


 

             To sum up, there are various types of misleading statistics such as sampling bias, misleading graphs, misleading advertisements and unfair comparisons. People should be aware of all kinds of statistics as they can be misleading in many cases. Therefore, precautions should be taken in order to avoid misinterpreting the data. As can be predicted, in the future, if misleading statistics still exist, people will lose their belief in statistics and this will cause a substantial impact on everything ranging from bankruptcy of companies to an economic crisis.

Thursday, March 20, 2014

Misleading Statistics (Term 1, 2014 Monday Wednesday Friday 11:30 AM Class)


                Nowadays, all over the world, numerous information has been put into statistics  most of which have the potential to influence people successfully. Statistics are collections or comparisons of data which represent not only facts but also measurements and in some cases, lies. These group writings are written in order to alert people not to make generalizations without further analysis. Statistics should be approached with caution: some provide relevant information whereas others are unreliable. First and foremost, this assignment will discuss about samples and polls then on how graphs, political statistics and advertisements can be misleading. Lastly, it contains some examples of false information.



Group 5:
(Word Count: 1093)
Level: EAP5
Class day/time: Mon/Wed/Fri 11:30 – 13:00
Samples
Nowadays, statistics is crucial for business, politics, education and science. It is also present in journals, reports, advertisements, and so on. However, not all statistics are reliable. Some are misused in order to convince citizens or customers. Figures catch audiences’ attention, but it does not mean that figures are always true and accurate. It is not very difficult to manipulate statistics. This essay emphasizes the importance of samples in statistics and polls. The sample size influences the accuracy of the results but size is only one point of the topic. There are two points more that we are going to study. One is sample type and another is categories.
To begin, it is important to know the types of samples which are used to make polls or statistics. Although, many types of samples are used but some of them are misused. There are two main types of samples, Probability Sampling and Non-Probability Sampling. Probability Sampling involves population sample selection, which means that everyone in the public has a chance to be selected, randomly, as a representative of the population (Crossmen, date unknown). For example, we go to interview the students in Sisowat High School to know what subjects they like to study, so we randomly select one or two students from each class. Most students, that we have selected, said they want to study math, so we simply assume that most students studying in Sisowat High School like to study math. Whereas, for non-probability sampling, people have no chance of being selected randomly and the selection based on people’s interest (Crossman, date unknown). For instance, a sport supervisor  want to interview people who like winter sport about the kind of winter sport they like to play, so he go to ask a group of people who like different kind of sports. In this case, he only want to interview people who interested in winter sport, so the other people in the group who does not interested in winter sport have no chance to be selected.

In addition, statistics is not reliable when the generalization is made from a smal lsample size. It means that a small sample size does not accurately represent the whole population. The reliability of generalizations made from small samples are weak (Rowntree, 1982). For instance, imagine if the Pizza Company wants to study the number of Cambodians who like to eat pizza and it also wants to represent in percentage to persuade that new generations really like to eat their pizza so it has to select a sample. If 100 students are chosen to be interviewed and if the result shows that 80 percent of Cambodia's people like to eat pizza. It will not be accurate because it does not show the number of the whole population. Furthermore, there is also a self-selection sample in some polls which is made the statistics undependable because there are some people who would like to vote and other people who do not like to participate in the vote ( Renka, 2010). For example, there was an election in 2000, Newsweek published the following poll results. It was publicized that Nader was the only candidate worth voting for, because he got the 9 percent of the vote so people expected him to have at least 9 percent of the vote. But at the end of the real vote, he had only 3 percent. In this case, it was because of the self-selected sample : which gave some people who are underage can vote and other people did not participate in this vote, maybe because of their work or because they thought that they would participate only in the real vote. Moreover, people who participated in this vote maybe they are Nader's supporters and they are interested in the questions.

Accurate statistics have to use samples that are representative of the population. Statisticians divide the sample in categories: age, sex, level of education, social class, and so on. An issue is different in a country where the majority of population is young (for example in Cambodia) or in a country where the number of elderly people is important (Japan or European countries). Recently, an article in The Guardian drew attention on the question of childcare in UK (Parents struggling with cost of childcare, 2014). The journalist provided three important numbers:
- A third of parents with children under five spent 30% of salary on childcare.
- 50 % spent at least 20%.
- 25% say they did not ask their family for help to do childminding.
The survey was concerned with 1625 people, an accurate number for a poll. However, the article reports that it comes from the website GoodCareGuide.co.uk. This website is a kind of TripAdvisor for childcare places. Parents can post admirations about a nursery, nanny or childminder. We can assume that people who answer the questions are particularly involved in the topic of childcare, which can make statistics misleading. Moreover the reader has no information on the categories of family: number of children, salary, social class, geographical origin and residence. All these data are important because statistics have to be a fair reflection of the real population. For instance, a family with 50 000 dollars income can spend 30% for childcare more easily than another family with 10 000 dollars. Likewise, if your parents or your sister live 400 kilometres away from you, you can not leave your children in their place for one day. When it comes to the immigrants, some of them have no parents or relatives in UK, so they cannot ask them for help. Another thing is the single mothers problem. Many mothers educate their children alone. For them, the childcare is very important, because they absolutely need to work. Unfortunately, the website GoodCareGuide.co.uk does not provide information about the details of the survey and the categories precisely involved with it. But a Guardian reader, after a quick reading, keeps just the figures in their mind.

To sum up, after looking further, we can see that sample can mislead people by using different techniques including types, sizes and categorizes. This can show that not only statistics such as advertisements or graphs are unreliable, but sometimes sample can also be unreliable and should be looked at with caution. In order to avoid being mislead by sample in polls or statistics, people should look through the information first and interpret the data before jumping to the conclusion.

















Bibliography :


·         Author unknown (2014). Parents struggling with cost of childcare, researchers find. The Guardian [online]. Available from:
·         Good Care Guide (Date unknown). Available from:
http://www.goodcareguide.co.uk [Accessed 9 February 2014]
·         Rowntree, D. (1982). Statistics without tears: a primer for non-mathematicians. Harmondsworth: Penguin.
https://archive.org/stream/StatisticsWithoutTears/Rowntree-StatisticsWithoutTears#page/n21/mode/1up [Accessed 10February 2014]
·         Russell D. Renka (2010). The Good, the Bad, and the Ugly of Public Opinion Polls.
http://cstl-cla.semo.edu/renka/Renka_papers/polls.htm. [Accessed 10 February 2014] 
·         Crossman, A. (date unknown). "Types Of Sampling Designs."About.comSociology. Available from: <http://sociology.about.com/od/Research/a/sampling-designs.htm>. [Accessed: 17 February 2014]










EAP 5
Days: Mon/Wed/Fri Time: 11:30-13:00

ôôôôôôôô¦Ã´Ã´Ã´Ã´Ã´Ã´Ã´Ã´

Fourth Draft

MISLEADING GRAPHS
            There are many ways to misrepresent data in statistics, but one of the most egregious ways is using graphs. Graphs depict data in which numbers are often believed to have an accurate proportion. However, there are some cases when graphs can be misleading. A misleading graph is a graph which shows inaccurate data unintentionally or intentionally by the designers, and the readers may make incorrect assumptions from the graphs(Witte et al, 2009, Internet). Various types of graphs can be used such as bar graphs, pie graphs, line graphs, pictographs, and other types of graphs. This essay will demonstrate why misleading graphs are created, how line graphs and pictographs can be misleading and how to avoid misinterpretation of data in those graphs.
Graphs constituted as misleading can be used for variety of reasons. The designers purposefully design the graphs to hide or exaggerate the proper interpretation of the graphs. Without careful analysis, readers will misinterpret the data and fall into their traps. However, in some cases, misleading graphs may be created accidentally by users who lack knowledge of numerical data or the graphing software, by the readers own misinterpretation of the data, or because the data is difficult to convey accurately. Misleading graphs are often found in false advertisements which are used by some corrupted companies to convince customers to buy their products. They are also useful in financial data and corporate reports (Witte et al, 2009, Internet).

The following graphs are selective examples to illustrate how graphs are misleading.

Figure 1:As you can see in the picture above, this statistic is talking about the productivity of fertilizer. Why is the first graphdifferent from the second one? (Haney, Internet)


A


 
B
  





(Source: http://www.calvin.edu/academic/economics/faculty/bios/HaneyDocs/page-58978255.html)

If we take a look at the figure 1 at the first sight, graph B will gain the most attraction. The graph increases sharply, which means that the company got high productivity. However, the difference between these two graphs is vertical axis. The company used big interval numbers to make it look more attractive. To create these kinds of graphs, the first thing is to make sure that all the information is true. Secondly, change the proportion of the horizontal axis or vertical axis to make it more attractive or jaw dropping. This kind of graph is called the Gee-Wiz graph. This method has already been used for about 70 years. (Haney, Internet)



Figure 2:In this second picture, the graph shows someone’s GPA (Grade Point Average).  Turning to the point, as you look at the human and brain picture and the graph, arethere any differences? (Haney, Internet)

C





D





(Source: http://www.calvin.edu/academic/economics/faculty/bios/HaneyDocs/page-58978264.html)

There is a small difference between 3.0 and 3.5 by bar graph. However, in the second graph, the writer converted the simple bar graph to pictograph by using different scales into two pictures. Picture C will get the most eye-catching because it is bigger than picture D. The strategy is to raise the head length and expand the width of the picture. The length and width can change the area of the picture. The formula is Area= Length X Width (Haney, Internet). People will get confused easily with the pictograph at the first sight because all of the pictures are creative, reflective and deceptive.(Haney, Internet)



Figure 3:This graph looks different from the others because it is 3D. Based on the graph, does it look misleading? How can this graph be reliable? (BBC, 2014, Internet)







(Source: http://www.bbc.co.uk/schools/gcsebitesize/maths/statistics/representingdata2rev5.shtml)

The figure 3 shows the number of singles sold from 1995 to 1998. The bar graph is different from the other graphs because it is used as a 3D bar chart to draw attention. There are some reasons which cause this bar graph to be misleading. First, there is no accurate scale for the vertical axis. Furthermore, the perspective can make a huge difference of the number of sales in 1995 comparing to other years. It would be better if the bar graph was made in 2D picture.(BBC, 2014, Internet)
Misleading graphs are commonly found when there is statistic presentation. To avoid misinterpretation, here are some key strategies to help avoid the misleading graphs. First of all, find out thetitle of the graph and then make sure what it is about. If it is a line or bar chart, label both axes. Notice that the scale of a line or bar chart must have the same intervals and start with zero. Figure 1 is a good example to clearly demonstrate how the misleading graph works in different scales. In picture A, the interval of the vertical axis is 10,starting from 0 while picture B has only one interval different and it does not start with 0. Consequently, it looks like picture B has a sharp rise. Actually, the two graphs come from the same statistics about productivity of fertilizers. It is easy to make the two graphs more accurate by equally numbering the scales of vertical axis.
The source of the data is another thing to consider. Also,make sure that the sourceis reliable. For pictographs, the size of a symbol must be uniforms. In figure 2, picture C does not show the uniform size with the picture D. The width of picture C is longer than picture D. To make the pictures accurate, both pictures should be of the same size. One more common type of misleading graphs is 3D bar charts which are used to easily cheat people. Figure 3 can be an excellent example which shows the sales for 1995 were far better than the other years. In fact, it is identical to sales in 1997. To deal with the 3D bar chart, scales for vertical axis must be numbered properly. (Faculty.atu.edu, Internet)
Graphs can be easily used to mislead people. By using some methods, people undoubtedly believe the misrepresentation of data in the graphs without much analysis. Nonetheless, it can be really difficult to differentiate misleading graphs sometimes. Still, people have to be very careful with every graph they encounter because they do notall provide reliable information.

ôôôôôôôô¦Ã´Ã´Ã´Ã´Ã´Ã´Ã´Ã´









BIBLIOGRAPHY
Author unknown,(2012).Misleading graph [online].BBC. Available from: http://www.bbc.co.uk/schools/gcsebitesize/maths/statistics/representingdata2rev5.shtm
[Accessed 10 February 2014].

Author unknown.Misleading Graphs and Statistics [online]. Available from:
http://faculty.atu.edu/mfinan/2043/section31.pdf
[Accessed 7 February 2014].

Haney, Becky. How to lie with statistics [online]. Calvin College. Available from:
http://www.calvin.edu/academic/economics/faculty/bios/ Haney Docs /page-58978264.html
[Accessed 10 February 2014].

Haney, Becky. How to lie with statistics [online]. Calvin College. Available from:
http://www.calvin.edu/academic/economics/faculty/bios/ Haney Docs /page-58978255.html
[Accessed 10 February 2014].

Witte, R. S., & Witte, J. S. (2009). Statistics 9th edition [online]. (9th ed.). New Jersey: John Wiley & Sons, Inc. Available from:
https://www.boundless.com/statistics/frequency-distributions/frequency-distributions-for-qualitative-data/misleading-graphs/
[Accessed 10 February 2014].






MISLEADING POLITICAL STATISTICS
    Statistics are the science of collecting numbers which represent a fact or measurement. Statistics have been used in advertisements, news, politics, science, and also economics, in the forms of tables, charts, bars, graphs, polls, and so on. Statistics can be used to indicate examinations of facts; however, in some cases, people create fake statistics which make people misunderstand it. In this essay, we will describe about a type of misleading statistics which is about politics and we will also talk about the causes and effects with some solutions along with it too.
    First of all, the causes of misleading political statistic would be desires of politicians to catch people’s interest and also journalists’ goals of catching people’s attention toward their articles or projects. Some politicians may use misleading statistics to reach their goals in some ways such as to promote themselves before the election or to convince citizens about the development of the country and the achievements during their time of becoming president. For example, the misleading political statistic which happened in the last two years in the United States that was provided by Obama about the unemployed veterans’ rate. It was announced that the percentage of the unemployment was lower after Obama entered the office comparing to the period before he was elected. In fact, the number of unemployment was already low at the time when he came into the office (Michael Tennart, unknown year). The statistic which was used by Obama can also be misleading because in this type of issue, political statistics can earn the interest of the citizens for the politicians as it shows the ability to solve problems and how successful the current president’s works are. Other than misleading statistics that were made by politicians, some other misleading statistics can be made by journalists or reporters as well. For instance, from what we have studied, there was an online poll about the people that would get the most votes in the election. Comparing this online poll result to the real election result, we found that there were huge differences. It was misleading because not all the citizens participated in the polls and only groups of people who were really interested or those who live in modern lifestyles would join it, which made this statistic unreliable.
    Secondly, as misleading political statistics mostly happen intentionally, it may widely affect citizens’ lives after they believed in political lies. Political statistics may be able to make people believe and make assumption quickly without careful analysis, however, for people who see further, they will think critically about the source of statistics which can help them avoid making wrong decisions, while those who may simply be persuaded by the statistics, would make wrong choices. For example, before every election, politicians from all parties will spread policies to attract supporters. Some politicians may stretch the truth by using statistics if they think it will increase the rate of voters. So some people may jump to conclusions quickly which makes them choose the wrong leader or party to lead their country. Other than this, misleading political statistics will also make people unsatisfied. Citizens need the leader who is competent and able to manage all the problems. As soon as the citizens are able to reveal the fake information or misleading data they will lose hope or belief and feel less confident about the leader’s policies (Richard R.Lau, Lee Singleman, Ivy Brown Rovner, 2007). Another example, politicians could promise that if they win the election, the government will pay for the citizens’ medical services. Thus, some people who are knowledgeable about economic status of the country will not believe in that kind of policies which is almost impossible to be provided. They would think that it is just a fake promise to persuade people to believe and vote for them. In addition to this, when the leader cannot complete what they promised, people will protest against the government by holding demonstrations or strikes. These can affect the social security and have an impact on the investment processes which may affect investors’ decisions as they might be worrying about conducting business in an insecure political country, which leads to another economic crisis in the country.
    Finally, the main solution to this problem is to clarify the statistics. The first suggested solution is to learn more about the methods of the survey. Generally, misleading political statistics are made by using graphs or asking biased questions to get the result that can benefit them (Harwell, 2010) or using amounts or number to exaggerate the result instead of using percentage. The second advice is to clarify the meaning of words included in the statistic. It occurs frequently that people misunderstand the definition of the word being used which is the statistics’ goal of making it misleading. Beside these advices above, it is suggested that readers should understand clearly about the mode, median, and mean average because different types of average are being used to provide different results. The last thing that is advised is to make sure that the statistic included reliable sources. Normally, the reliable sources would be provided below the statistic by using an asterisk as the sign for the sources in small letters.
    In conclusion, misleading political statistics can be found everywhere. People try to produce misleading statistics for different kinds of benefits so people should try to avoid being convinced by any kind of misleading statistics. For some reason, people should not always believe what they see in statistics; instead they should compare it with real life or situations. It would be more helpful to identify the true and the fake information using reality. This is also the key to avoid misleading statistic.




Bibliography
Michael Tennart. Obama’s Misleading Statistic on Veterans’ Unemployment. Retrieved from http://www.thenewamerican.com/usnews/politics/item/13389-obama%E2%80%99s-misleading-statistic-on-veterans%E2%80%99-unemployment
Harwell. (2010, 03). How to create Misleading statistics in 6 easy steps.  http://blog.makingitclear.com/2010/03/25/statistics/
Richard R. Lau, Lee Singlman, Ivy Brown Rovner (2007). The Effect of Negative Political Campaigns: A Meta-Analytic Reassessment. Retrieve from www.web.posc.jmu/polbehavior/readings/Topic10_Median/lau+singleman+rovner%.com.
Best,



Misleading Advertisements
In addition to the issues, which were mentioned earlier, let us take a look at how advertisements could be misleading with and without statistics.
The most common ways that have been used by marketers are the use of misleading statistics to link up with the terms of “more”, ”new”, ”discount”, and fear factors which are used to threaten people in their advertisements (False advertising, 2012, Internet). In this technique, the marketers might use inappropriate samples to misinform the statistics, which leads to misleading advertisements. One good example of this technique would be from the Colgate advertisement in 2007 when the company claimed that 80 percent of dentists recommended the brand. At that time, customers might conclude that Colgate was the best choice because only 20 percent of dentists recommended different brands. However, according to the Advertising Standards Authority, Colgate’s survey was only based on telephone surveys of dentists and hygienist which were made by an independent market research company. Therefore, it turns out that the marketers just used a small sample size to make misleading statistics and included it in the advertisement (Derbyshire, 2007, Internet). Moreover, fear factors are used in advertisements to entice customers as well. Normally, they insert threatening words in their commercials to convince customers to assume that they need to purchase those goods. A good example would be the advertisement of ACME life insurance. It indicated that accidental and premature events could occur to the head ofAmerica’s families who live to age 75 on average, and leave no support to the family members. In this case, among the list of fearful things, the fear of death ranks at the top (Critical Reading of Advertising Material, 1996, Internet).
In addition, in some commercials, the producers not only like to conceal the real benefits of the products, but also paying the actors to mention the terms and conditions in inappropriate accents. To conceal the qualities means the creators try to persuade consumers that their products’ quality and appearance are the best while the actual products are usually in average conditions and forms. For instance, Activia yogurt and DanActive dairy drink advertised that it had been scientifically and clinically proven to regulate digestion and boost immunity system 10 percent better than the others. Nevertheless, the company was sued to pay $45 million and changed its health claim because their products have not been medically proven as the company claimed it would be (Troy, 2010, Internet). Furthermore, in some TV commercials, the actors are paid to mention about products’ terms and conditions in uncertain accents or accelerate their speech so that most customers could not understand. For example, in 1981, Moschitta, who was best known for rapid speech delivery, was paid to advertise in FedEx’s commercial. In this advertisement, it wrote “When it absolutely, positively has to be there overnight” in very big capital while Moschitta mentioned “Areas served, delivery times, and liability subject to limitations in our Service Guide” for just 5 seconds after it (John Moschitta, Jr, 2009, Internet).
Instead of inserting false statistics, there are two other popular ways which are used in advertisements to manipulate consumers. Firstly, and commonly, the producers hire celebrities to advertise their products and then follow by the use of fantasy words in the advertisements.Tobegin, using celebrities to advertise is a very effective marketing trap. It provides lots of advantages to manufacturers of the products because celebrities often generate attention from people. In this case, people might fall for the trap and purchase the products unconsciously. People may think those products must have high qualities since they are recommended by celebrities. Moreover, people would also think that by using those goods, it will make them beautiful like celebrities because there are some people out there who think “If the product is good enough for her, it’s good enough for me.” Additionally, it is obvious that obsessed fans would purchase products that are advertised by their idols. For instance, Wonderstruck perfume has rapidly become very popular after Taylor Swift advertised it. Thus, there is a high chance that people would be influenced to the products without examining it effectively, especially with celebrities’ presences (Suttle, 2014, Internet). Another persuasive marketing technique that is used by marketers is advertising dreams. They often use the glitz and glamour words to hide the impossible and unachievable result of their products. For example, nanoblur claimed that by using this cream, it could make you look younger in 40 seconds. It is obvious that they are just selling an impossible dream because even though the particles are at 700 nanometers, it is still too big to slip into the skin. Hence, they could only reflect the light from the skin’s surface (Food for thought, 2009, Internet).













Bibliography

            Simon, R., (2007). Critical Reading of Advertising Material [online]. 6 University Way, Bellevue Heights, SA, 5050. Available from:
hptt://www.rocketreader.com/newsletter/newsletter14.html
[Accessed 15 February 2014]
Troy, M., (2010). Dannon to Pay $45M to Settle Yogurt Lawsuit [online]. Available from:
hptt://abcnews.go.com/Business.dannon-settles-lawsuit/story?id=9950269
[Accessed 15 February 2014]
Author unknown, (2011). False advertising [online]. Available from:
hptt://en.wikipedia.org/wiki/False_advertising
[Accessed 15 February 2014]
Suttle, R., (2014). What are five advantages of using celebrities in advertising? [online]. Available from: hptt:// smallbusiness.chron.com/five-advantages-using-celebrities-advertising-34394.html. [Accessed 15 February 2014]
Derbyshire, D., (2007). Colgate gets the brush off for “misleading”ads [online]. Available from:
hptt://www.telegraph.co.uk/news/uknews/1539715/Colgate-gets-the-brush-off-for-misleading-ads.html
[Accessed 02 March 2014]
Author unknown, (2009). John Moschitta, Jr. [online]. Available from:
hptt:// en.wikipedia.org/wiki/John_Moschitto,_Jr.
[Accessed 02 March 2014]
Author unknown, (2009), Food for thought [online]. Available from:
hptt://flawlessgl.wordpress.com/tag/misleading-advertising/
[Accessed 02 March 2014]



Word Count: 852
False information impacts on statistics
For today’s world, technology has improved and found ways to describe figures or interests such as communication, descriptions, commercials and statistics. Statistics is a method used to organize data (Rowntree, 1982). Statistics can illustrate both lie and truth to the users so that people should be careful with what they are reading or creating. This essay will focus on false information and discuss about benefits and drawbacks of statistics as well as some samples.
Statistics are ubiquitous, especially in newspapers and commercials, and they are the main source of countless numbers of articles (Bain, 2011). For instance, there are advertisements about business, politics, games and television commercials. For example, a company tried to convince customers to buy their products by stating that many scientists recommended their products. Some individuals instantly believed it without any further analysis. Experts might value the product, but it did not specify the exact number of scientists or whether those scientists were qualified to endorse that product or not. Vishal, resident of Gurgaon in India, tried to copy a commercial “Red bull gives you wing’’. Vishal who drank thirty cans of red bull jumped off a building due to that commercial. The commercial indicated that after having the energetic drink, anyone would be able to grow wings and lift off their feet from the ground (Virginia, 2010).



In another commercial “All natural, All cola” from Red bull, the narrator said that Red bull drinks contained only natural ingredients, but later there were studies of consumers suffering from dehydration, tremors, heat strokes and heart attacks (Moefu, 2010). Why was this information not included in the commercial? This was done in order to make the commercial become beneficial for the company, then people could understand how good Red bull was, so the writer chose what was suitable to present. Similarly, some companies benefited by creating the experiments with a small group of people, which proved that their products have good quality (Korn, 2009).
Nonetheless, although statistics can show parts of the truth, they also cause misinterpretations to the users (Korn, 2009). An Internet advertisement shows that “wrinkles free in only minutes”. What is wrong with this phrase? The “only” and “minutes” are obviously created to motivate readers to use their products. The attraction of “no wrinkles” in any “minutes” was used in this case, but we do not know the real amount of minutes (Ryan 2008). For this reason, the company might be able to boost their popularity and attraction when users read the products’ tags.
The majority of statistics are used for describing numbers, figures, amounts and percentages. Specifically, data can help the users to classify things and identify the problems (Rowntree, 1982). Coloration also plays a salient role in some misleading statistics because it tends to confuse readers about what it really tries to show (Robbins, 2012). The coloring parts in different charts, which are on the same topic, do not even connect well with each other. Then the statistics could be hardly true. All the mistakes would only fool people and convince readers to agree with their confused theory (Robbins, 2012). The following picture is one of the graph charts that exist with the mistake above.
The colors in the map and the accompanying pie chart are different. For this reason, on the map (right), we see that upper-black color was used to represent Kentucky, and then the readers might make an assumption that the black color on the pie chart (left) would also refer to Kentucky without reading the labels carefully. Actually, the black color on the pie chart (left) indicates Piedmont. So, consistency of color is very important in graphs showing the same data of one statistic (Robbins, 2012). People tend to get a better understanding in comparing two or more tables or graphs on the topic by reading through those statistics (Pinkmonkey, Study Guide, 2011). However, in some cases like picture above, the charts are not even accurate so readers have to read carefully in order to avoid the wrong statistics.
Moreover, there are many ways that graph can be misleading such as using casual images, reversing skills, and hiding small samples (Post, 2011). Firstly, attractive images could bewilder people about the graphs. For example, this particular company was trying to decide which products had given them more profits than the others. This graph (below) was made with little pictures of dogs, cats and food cans. That was an example of bad sampling that readers cannot understand just by reading the graph (Post, 2011). Whenever the statistics are fabricated or intentionally invented without proof, many people will be misled with the data that has been shown (Korn, 2009).

To sum up, statistics are widely used. There is clear evidence that shows statistics can coat the truth too. Statistics are essential in daily life, but they are also notorious for the purpose of making false and misleading arguments (Korn, 2009). As a result, readers in particular areas should search more even if they have the statistics since they can also be deceiving when the truth is being manipulated or hidden for specific purposes (Can, 2011).
Bibliography
·         Tuesday, M. (05, March 2009). How statistics can be manipulated. Retrieved from http://www.helium.com/items/1364782-statistics [Accessed 27 May 2012].
·         Bain, S. (17, February 2009). How statistics can be manipulated. Retrieved from http://www.helium.com/items/1344433-statistics-media-media-manipulation [Accessed 10 Feb 2014].
·         Tran, Can. (15, March 2009). How statistics can be manipulated. Retrieved from http://www.helium.com/items/1377267-statistics [Accessed on 10 Feb 2014]
·         Korn, B. (2009). Responsible thinking: Principles. Misleading Statistics. Retrieved from http://www.truthpizza.org/logic/stats.htm . [Accessed 12 Feb 2012].
·         Pinkmonkey, (2011). Study Guide on misleading statistics: Pinkmonkey. Retrieved from http://www.pinkmonkey.com/studyguides/subjects/stats/chap1/s0101201.asp [Accessed 12 Feb 2012].
·         Rowntree.D, Statistics without Tear. Penguin (1982), pp 14-21. [Accessed on 9 Feb 2014]
·         Australians Competition & Consumer Commission. False or misleading claims. (n.d.). Retrieved from http://www.accc.gov.au/consumers/misleading-claims-advertising/false-or-misleading-claims [Accessed on 10 Feb 2014]
·         Virginia, M. (2010, March 14). Lack of information causes negative impact. Retrieved from http://lackofinformationnegativeimpact.blogspot.com [Accessed on 8 Feb 2014]
·         Dahiya, H., & Saxena, S. (2013, May 12). Hoping to fly to india gate, man jumps from building after drinking red bull. Retrieved from http://newsthatmattersnot.com/news/man-jumps-from-building-after-drinking-red-bull [Accessed on 10 Feb 2014]
·         Robbins, N. (2012, Sep 19) Misleading Groups of Charts. Forbes, Retrieved from http://www.forbes.com/sites/naomirobbins/2012/09/19/misleading-groups-of-charts/ [Accessed on 8 Feb 2014]
·         Post, J. (2011).  Are graphs misleading? Retrieved from https://www.jerrypost.com/MIS/Discussions/Discuss02.html [Accessed on 10 Feb 2014]
·         Ryan. (2008, August 15). Dishonesty in advertising: A little bit goes a long way. Retrieved from http://baronandcompany.wordpress.com/2008/08/15/dishonesty-in-advertising-a-little-bit-goes-a-long-way/ [Accessed on 7 Feb 2014]



            In conclusion, from what has written above, it appears that there are so many kinds of tricks which are used to make statistics misleading. Indeed, statistics are not always true or reliable even though they appear to be scientific. That is why people should be careful when encountering graphs, advertisements, samples, information in the media and political polls. In fact, the reader should find more sources and do some research before trusting the given information in statistics. Since the societies are full of information from the media, we ought to make accurate generalizations after thinking critically.