Sunday, December 8, 2013

Misleading Statistics (Monday Wednesday Friday Class)


                Statistics are being commonly used in daily life. All over the world, they play important roles in many fields such as politics, business, science, education, medication and so on. Statistics is the study of collection, organization, analysis, interpretation and presentation of data. Statistics should be interpreted with caution as they can be misleading; they can both tell lie or tell the truth. This essay will cover main five areas, including intentionally or unintentionally, misleading statistics, false advertisements, reliable and unreliable surveys, averages and pictographs.




“Statistics should be interpreted with caution as they can be misleading; they can both lie and tell the truth”
4th Draft:Reliable and Unreliable Survey
Statistics are being commonly used in daily life. All over the world, they play important roles in many fields such as politics, business, science, education, medication and so on. Statistics is the study of collection, organization, analysis, interpretation and presentation of data. Statistics should be interpreted with caution as they can be misleading; they can both tell lie or tell the truth. This essay will cover main five areas, including intentionally or unintentionally, misleading statistics, false advertisements, reliable and unreliable surveys, averages and pictographs.
Survey is the process of gathering people’s perception usually known as respondents’ ideas in certain research purposes. There are two main approaches which can be used to conduct a survey including interview and questionnaire. Both approaches can either be conducted through phonecall or the internet as well as face to face. Surprisingly, surveys can be misleading in case that, they are not conducted properly or sometimes are being asked about a topic that respondents tend to give untrue answers especially about self-reporting(Haney, 2013, Internet).
However, not everysurveycan be assumed to be unreliable information, by looking at the way how data is received. Hence, there are three essential factors which affect the reliability of survey including length (total number of questions), the size of sample group and the quality of questions (Brown n.d.). If a questionnaire is reasonably long and questions are designed fairly without bias, it should be reliable. Moreover, the most reliable survey questions are open ended questions rather than multiple choice which allow respondent to answer freely with their own answers.Lastly, size ofsample group which is sensibly acceptable and may represent the whole populationwill also give an accurate data for users.
In addition, surveys which are conducted with professionalism are also reliable.Reputation of surveyors’ entity also playsa key role in making the reliability of survey as well,for example surveys are done by United Nation, IMF, World Bank and Asian Development Bank so on.
Furthermore, it is true that most statistics are not reliable and can be misused badly sometimes. However, on the bright side, statistics can provide people with vital information if theyuse them correctly. To know if a statistic is reliable or not, you need to know its source. Knowing its source helps people a lot because it will give them an accurate statistic and the right information. Take two companies, for example company A makes up their statistics on product sales occasionally. In contrast, company B rarely makes up their statistic. This shows that statistics from company A are more reliable than B due to its frequent checking and the reliable source behind them.
In addition, to get a good and an accurate statistic, you need to understand well the statistic you are reading. Sometimes, it is your understanding that can either lead to a misleading statistic or a reliable one, for instance a statistic full of numbers that requires your mathematic knowledge to figure out what it really means. However, if you are not familiar with the math you can get the wrong information from the statistic. Sometimes statistics are reliable; however, because of our misunderstanding, we can convert them from reliable to unreliable. Additionally, understanding the procedures that are used to create each statistics is also very important because this can help you to decide whether you think the statistics are reliable or not. Knowing how they make a statistic can be very helpful because you can either choose to believe it or not to believe it. It is difficult to see through statistics with just your two eyes, but it can be very easy to get fool by statistics because they can be very persuasive and convincing.
On the other hand, there are so many ways to mislead people by surveys. Generally, the surveyors try to collect the answer and opinion from the surveys in order to make people believe them and make profit from it. So it is considered as an unreliable survey. To begin with, survey can be misleading when the sample size is small or it does not represent the whole population. If you ask 100 people, for example, if they approve or disapprove of discrimination of homosexuals, and 55 of them say they approve, you might assume that about 55% of the whole population approve. However, this result just happened by chance to interview an unusually large percentage of people who approve (Lies, Damned Lies, and Statistic, 2009, Internet).Since, this may happen because without being aware of it, you may select less religious conservative’s people who disapprove than those who approve in the society. This is the problem of sample size. The smaller the sample size, the greater the influence of luck on the results you get (Lies, Damned Lies, and Statistic, 2009, Internet).Asking for opinion from a group of people, and make a generalization as representative of the whole population cannot be reliable at all.
Additionally, one reason that unreliable survey occur recently is through technology such as mobile phone and the internet. For example, surveys conducted by text messaging can easily mislead people because the answer will only get from respondents who have cellphones. The results that they receive from the surveys are just a small percentage only (Haney, 2013, Internet). As a result, it is difficult to have a reliable survey and it is also true for the internet surveys.
Another probable reason is mostly surveyors come up with personal questions when they randomly select people, they would like to ask but, he or she might feel embarrassed to answer. So the answers are often different from what they want to know (Haney, 2013, Internet).  For instance, if a company wants to advertise their toilet tissues and they could ask people “what kind of tissues do people like to use?” People will avoid answering or they will just say they use water instead of tissues.
Last but not least, biased questions is another way of making surveys become unreliable and it is commonly in a form of multiple choice questions. For that multiple choice questions people cannot show their own opinion or answer because all the answers that the surveyor want people to choose are already prepared. Therefore, they can mislead them by designed questions thatthey wanted respondents to answer (Haney, 2013, Internet). 
            Everyone has experienced the survey but they just do not know whether the surveys they read are true or not. So, there are plenty of methods to fight against unreliable survey. First of all, one way to avoid those unreliable surveys is look at the style of the survey. It is better if questionnaires are designed in an open-ended survey question instead of multiple choice. The reason we should do that is because this kind of survey question allows people to view their answers and opinions. For example, if an interviewer asks a question, they leave out a space for the participants to answer. This is considered to be a reliable survey. Another method to fight with unreliable surveys is when you read a report or an article, they could provide the number of something. However, the reporters usually use the word “average” or give us the data with percentages. This could make us confused. So, the reader should trust a report with specific numbers (Haney, 2013, Internet). 
All in all, statistics can both tell lies or truths mainly based on their sources which are surveys. Hence, surveys should be conducted properly with professionalism and independence. However, some surveyors try to trick people by many ways in order to get benefit from the statistic that they made. Therefore, as a reader of statistics we should not trust and finalize any data that is given by surveys without careful evaluation unless you know the sources and the proofs behind them.



Bibliography
1. Pingback: Lies, Damned Lies, and Statistics (10): How (Not) to Frame Survey Questions « P.A.P. Blog – Politics, Art and Philosophy
“Lies, Damned Lies, and Statistics: Too Small Sample Sizes in Surveys”
2. Brown, James Dean. “Reliability of Surveys.”Statistics Corner, n.d.:18-21
3. Haney, Dr. Calvin College.n.d. http//www.calvin.edu/academic/economics/faculty/bios/HaneyDocs/
(accessed September 10, 2013).









Intentionally and Unintentionally Misleading Statistics
Statistics refer to a fact or piece of data obtained from a study of a large quantity of numerical data. Statistics are commonly misleading, and those misleading statistics are divided in to two types: intentionally misleading statistics and unintentionally misleading statistics.
Intentionally Misleading Statistics
Intentionally misleading statistics refer to a presentation of wrongresearch data which in turn gives wrong guidance in favor to a certain situation(Misleading Statistics Definition?, 2013, Internet). People use this type of statistics to trick and to illustrate other people to believe what they are saying are true.Intentionally misleading statistics can appear in many factors likepolitics as well as business.
Intentionally Misleading Statistics in Political Field
Sometimes politicians use this kind of statistic for the benefit of political conditions. They tend to create misleading statistics in order to gain popularity for their own party. They may use it in two main ways.
Pentagon: Intentionally and Unintentionally Misleading Statistics

Firstly, politicians tend to trick people for their party’s goodness. For a hypothetic example, Cambodian’s government party used to claim that they were working against poverty among Cambodians. However, they tended to show their success by helping to improve the living condition of those who were slightly below the standard; they ignored those who were far behind the living standard. As a result, the government claimed a great success of reducing poverty throughout the country. This is an intentionally misleading statistic due to the fact that there are still a lot more people who are living in severe poverty.Also, the opposing party in Cambodia was trying to claim their popularity among people through Facebook social network. This is misleading since not all Facebook users are able to vote, some of them may not be Cambodians, and some others are too young to vote.
Secondly, besides using misleading statistics to show off their good characteristics, politicians often use them to criticize the other party. Opposing parties often try to make people believe that the ruling party is not capable of governing the country by using misleading statistics such as pictographs (a pictorial symbol for a word or phrase) to illustrate the number of corruption cases, deforestations, land disputes, and poverty in Cambodia.

Fig. 1 “Chesterfield” cigarette
Intentionally Misleading Statistics in the Business
While statistics have been used to display numbers or data, businessmen and businesswomen make use of it through their advertisements in order to gain more benefits from selling their products. However, most of the statistics taken from the advertisements are intentionally misleading statistics. Since business is always about making profits, business people always play with the statistics to overwhelm the customers. In terms of advertising, intentionally misleading statistics can be caused by many errors such as unreliable information and inaccurate numbers or percentages, and so on.
Most of the advertisements confuse the customers with unreliable information. For instance, there was an advertisement of cigarettes called “Chesterfield” (Figure 1). The advertiser tried to promote his product by saying that there were no negative side effects on the nose, throat, and sinuses of the group whowere smoking Chesterfield. However, he did not mention about the lungs at all. Actually, cigarettes would produce bad effectsdirectlyto the lungs; and many smokers have faced lung cancer and death.
The majority of the advertisers may trick the customers with the numbers or percentages. To cite another example, there is a title from an advertisement “3 Out of 4 Doctors Would Recommend Flora Pro-Activ Mini Drink” (Figure 2). This advertisement attempts to grab the customers’ attention by providing the number 3 out of 4 or 75% out of 100% of the doctors would recommend drinking Flora Pro-Activ. In this case, the advertisers try to make a Wow Factor. Wow Factor means making the number looking bigger than what it actually is by using numbers or percentages (Increasing the “wow factor”, 2013, Internet). However, according to the article, there are less than 150 doctors who would suggest to people who want to lower their weight to drink it. So, only 150 doctors are considered as 100%. This is just a small and limited number compared to the total doctors in the world.
Unintentionally Misleading Statistics

Fig. 2 “Flora Pro-Activ” mini drink
While most of the misleading statistics are intentional, there are also some unintentionally misleading statistics. Unintentionally misleading statistics refer to an unintentional error of a presentation of wrong research data which in turn gives wrong explanation in favor to a certain situation. These misleading statistics are unintentional because sometimes they are lack of further information on the statistics orhave poor system of statistics analysis.
For example, the average SAT scores in 1998, North Dakota had higher average SAT scores than New Jersey, although New Jersey ranked 2nd in spending and North Dakota ranked 45th. This explainsthe different systems that the two states were using (Alden, 2005-7, Internet). It was an unintentionally misleading statistic because they did not give out full information about those differentsystems for taking the tests in those different states.


Fig. 3 Average SAT Scores, 1998



Fig. 4 Mandatory and Voluntary helmet use
The next example would be the number of offences recorded and categorized by the police as the “most serious violence against the person” between April and June in 2008 rose by 22% in Guardian News. This statistic showed the poor system analysis in which due to the change in interpretation of the counting rules. The older rule was to only count the number of violence against the person in which resulted in injury. While the newer rule categorized the violence against the person with or without injury as long as the violence is intentional (Pallant, 2009, 34).This though has increased the number of violence in just three months.
Another example is about mandatory and voluntary helmet use (Alden, 2005-7, Internet). This statistics indicates that although the reports come from the government, it does not mean that they do not have any error. The errors of this statistic are those about comparing different kinds of areas, situations, or traffics of helmet use, which cause confusion to the readers and mislead them to say that using helmets is more dangerous than not using a helmet.



Bibliography:
1.      Author unknown, (2013).  Misleading Statistics Definition? [online]. Ask.com.
[Accessed 7 November 2013]
2.      Author unknown, (2013). Increasing the “wow factor” [online]. Calvin College. Available from: http://www.calvin.edu/academic/economics/faculty/bios/HaneyDocs/page-58978255.html [Accessed 9 November 2013]
3.      Lori Alden, (2005-7). Statistics can be misleading [online].
[Accessed 7 November 2013]
4.      Pentagon: Intentionally and Unintentionally Misleading Statistics

Anne Pallant, (2009).Surge in violence, or just a quirk?. Garnet Publishing Ltd.: UK.




MISLEADING ADVERTISEMENTS
We are surrounded by persuasive advertisements and commercials everyday, almost everywhere, on the Internet, big banners, posters along the roads and probably the thing we spend the most time with, television. Although those advertisements seem very attractive, they can be really deceiving. Since we all want our money’s worth of what we are purchasing, we should be more cautiously interpreting advertisements through some common tricky methods marketers use.
One dishonest advertising technique to attract customers is “bait and switch”. Marketers advertise a product (the bait) in a very attractive price which is explained as out of stock or unavailable once the customer comes to the store to buy the product. He or she is then encouraged to purchase a similar product with a more expensive price (the switching process). One common example of the practice in this technique is by offering a free ink cartridge (bait) within the purchase of a printer (Arnold). However, when the consumer arrives the store, he or she is told that the cartridge supplies are limited and are all given out. The consumer is then convinced to buy one cartridge instead (switch).
 Merchants can also impress people that their product is the best by using the inconsistent comparison method. Advertisers only compare their item with others that they superior to, leaving out products that are better than theirs. An advertisement might say their product is cheaper than product A and is better in quality than product B. People who see this advertisement might think that this product is the best among all products in all ways whereas in fact product A is the most expensive and product B is the lowest in quality, or maybe there are other non-included attributes where the advertised product may not be as good as others.
Another deceptive excuse to shoot up the price of a product is by adding fillers. Often found in food, fillers are cheap non-nutritive ingredients which are added in order to gain the overall weight and trick consumers into thinking that they are buying food with lots of beneficial ingredients at a reasonable price. For example, 15% of broth is injected into meat which decreases the amount of meat to 85%. This costs producers less than not adding fillers and having 100% meat.
Angel dusting is also seen as a common marketing practice in misleading advertisements. This method misleads consumer’s expectation of gaining benefits from products such as cosmetic, dietary supplement or food product because of some helpful ingredients included. However, the amount of those ingredients added are insufficient, thus will barely make any difference from not being added. Take a cereal claiming that it contains “12 essential vitamins and minerals” as an example. It is true that these vitamins are helpful and are contained in the cereal but what consumers do not notice about it is that each ingredient added is only 1% or less providing no measurable beneficial effects.
Customers can also be tricked by pictures they see from the advertisements. The practice in this method is called misleading illustrations. Pictures on packages of food can show additional ingredients which are not really included in what they are selling (Poulter, 2010). Another way of using images to deceive people is through showing impressive good-looking photos of food on advertisements. In reality, it might not be the same size and as fresh looking. As an example in this case, Burger Kings company’s commercial on television was showing a picture of their Tendercrisp burger that looks fresh and large with good looking ingredients seen from the side. However, when people go to buy the burger from the store, they appear smaller, thinner and the ingredients are not as visible from the side.
Another form of promoting a product to seem to be the best is the “no risks” method. For example, if asked to choose between a drug that can decrease your chances of getting cancer by 50 percent or a drug that rids only cancer for one out of the hundreds who are in the experiment, you would choose the drug that lowers chances of getting cancer without any hesitation. 50% and 1 out of a hundred are just emphasized numbers used to draw consumers into buying their medicine (Mike Adams). How it works is that in the experiment involving 100 people, there are 2 people that are more likely to get cancer than the rest. Those 2 people then took the same medicine and the results came out that only one of the two got cancer, meaning that the drug works but not on both, but it still works. Though only one out of two were tested and the drug only worked on one person, the pharmaceutical industry will make a claim that the use of this drug will reduce the person’s risks of getting cancer by 50%. This then will be exaggerated even more when it hits the headlines of the newspaper, emphasizing nothing but the benefits of the drugs whilst minimizing the risks and the side effects of the drug.
Last but not least, the “guarantee without a remedy specified” method is also used to attract customers. Every purchase of a product from a company comes with a guarantee slip. It is a form or a contract to prove that you have bought the product from the company and that your product is guaranteed to be renewed if it is broken or you are not satisfied with the product. This is just a simple trick that gets people all the time. The trick is in the guarantee slip. Some companies make these slips about 3 pages long talking about all sorts of stuff because they know that the consumer would get bored reading the first three lines of the paper. Not knowing fully what they are guaranteed to, you are being tricked and probably scammed. An example of this would be, when people actually return their products, they will then find out that they have to pay for the shipping fee or maybe they will get only 70% of the full price back. These conditions are written in the guarantee slip, somewhere that the customers haven’t yet to read.

BIBLIOGRAPHY
1. False advertising. [online]. Retrieved from: http://en.wikipedia.org/wiki/False_advertising [Oct 21 2013]
2. Poulter, S. (2010, July 21). Advertising watchdog finds burger king guilty of telling a whopper over serving size. [online]. Retrieved from: http://www.dailymail.co.uk/news/article-1296485/Advertising-watchdog-finds-Burger-King-guilty-telling-whopper-serving-size.html
3. Anderson, A. & Demand Media. Examples of Bait and Switch advertising. [online] Retrieved from:  http://smallbusiness.chron.com/examples-bait-switch-advertising-10575.html
4. Mike, A. (n.d.). Lying with statistics: How conventional medicine confuses the public with absolute risk vs. relative risk. Retrieved from http://www.breastcancerchoices.org/rr.html











EAP 5
Group 2:
Average is a type of statistics that people, companies and organizations, use to mislead others. Therefore, what is an average? Average has different meanings in different circumstances. One of its definitions is to talk about people’s status, quality or ability. However, in statistics, average is defined as the result obtained by adding several quantities together and then dividing this total by the number of quantities (Longman dictionary of contemporary English, 2010).
                In statistics, average has 3 different meanings; Mean, Median, and Mode. A mean average is the sum of a collection of numbers divided by the size of the sample. For instance, in a group of people with the age of 16, 16, 17 and 19, their mean average is 17. To get this result, we add up all the numbers (16+16+17+19) then divide the sum by the size of the sample which is 4. Thus, we will get the outcome of the mean average which is 17. Despite using mathematics to calculate the amount, we can take the middle number as the average as well. This method is called the median average. As an example, we will use the previous figures again to clarify this type of average. In that case, the median average of the group of people is 17.  By putting all the numbers in a row with the smallest digit on the left and the largest digit on the right (16,16,17,19), we will see that 16 and 17 are the middle numbers. However, 16 is used twice so we will use that number once only. And as a result, the middle number of this statistics is 17, so 17 is the median average. On the other hand, mode average is the value that appears most frequently in a set of data. For instance, imagine your national test scores are 50, 50, 61, and 78. Hence, your mode average will be 50 because the number 50 appears twice as often as the other numbers (Chapter 2: Deceptively Mean, Internet).
                Most companies or organizations try to mislead people by using average statistics, which are technically true, to create a misleading situation. Because people trust the type of average that is being used, when the average statistics are shown by those companies, people usually fall for their tricks. Plus, all types of average are neither correct nor incorrect. All of these types are used in everyday life for the purpose of making people believe in a piece of information, and they can be misleading if people do not consider wisely.
For example, when universities want to advertise their school, they will always make an announcement about students who attain a Master’s degree with average earnings of $46,269 annually. On average, students who graduated with a Master’s degree, are able to earn a lot more money than those who graduated with a Bachelor’s degree. However, this statistics is misleading because those universities use the median average. This does not modify that after these Master’s degree students graduate from their school, they will earn the same salary as the data shows. As it depends on their intelligence and effort in their studying as well, which means that students who are hard-working and clever, might earn this much. On the contrary, lazy students might not. Therefore, other variables may also be influencing the result (Alden, 2005, Internet).
Another example, when people want to buy a house, they will always ask real-estate agents about its location and neighborhood’s income. This is a golden opportunity for the real-estate agents to take benefits from them. By just using different types of average statistics, they can make it misleading to people to persuade them to buy more houses at a really high price. Using the highest average among the 3 types can really help the real-estate agents to mislead those buyers about the income which these neighbors could bring in yearly. Then the customers will want to come and live in (Discover How to Make Money Online and Incomes, Internet). Thus, making use of the average is not always accurate, as once the data is misleading, it can lead people to a misunderstanding of the data.
On the other hand, Pretty picture or Pictograph is a type of chart and graph which is used to represent data by using pictures, symbols or icons. It represents different data in which it tells the number that each picture or symbol illustrates. Users can create a visual display of the information from this type of graph in which they can use to advertise their products, and to make the consumers feel more likely to buy the products as well. Most people do like to use pictographs to convey their statistics because it can provide an uncomplicated way to display the data. They are easy to read, understand and are fun to use. (Freebern, 2001, Internet)
In the world of statistics, there are many types of statistics that people use every day, but it does not mean that those statistics are always accurate and reliable. As for pictographs, if we do not use it correctly, the pictures themselves can distort data. When we look at the pictographs our eyes generally focus not only on the pictures but also focus on the area around the pictures as well (Chapter 6: Pretty Pictures, Internet). Therefore, what makes pictograph misleading?
Take a look at the pictures below:
                                                                                                                  
                                             
Fig. 1
     Source: http://www.calvin.edu/academic/economics/faculty/bios/HeneyDocs/page-59480389.hmtl
                                                                                               






This pictograph is basically a bar graph which is made up of a picture of Halloween candy. It repeats the number of candies to replicate the bar of data (See figure 1).
                                                                                                                                   
                                               
fig 2

Source: http://www.calvin.edu/academic/economics/faculty/bios/HeneyDocs/page-59480389.hmtl
                                                                                    
                                                                                                                                   








                                However, this pictograph is misleading when the picture of Candy corns are displayed in two dimensions, width as well as height (See figure 2). The pictograph shows that Michael’s candy is twice as high and twice as wide. Generally, it looks like the candy Michael collected is four times the amount of Shayna’s candy. (A Wrong Way and A Right Way, Internet). (See figure 2)
Here’s another example of pictograph, look at figure 3 and figure 4 below.
Fig 3



fig 4

Source: http://www.calvin.edu/academic/economics/faculty/bios/HaneyDocs/page-59448244.html

              
                                                                                                                                                                                                               

                                                                                                                                                                  


                      The bar chart and graph show the quantity of trash from 1960 to 2000. In figure 3, the trash cans increase significantly in 40 years and the width size expands larger and larger. However, it could be misleading because of using pictures that are not compatible with the numbers. In figure 4, it is more exact because in each chart they keep the same size and only make a change of its height so people cannot be confused with this graph (Misleading Graph #2, Internet). Therefore, if authors want to create misleading statistics, they can use this kind of method to the fullest.














Bibliography:

Longman dictionary of contemporary English. (2010).

Author unknown, . Chapter 2: Deceptively Mean [online]. Available from: http://www.calvin.edu/academic/economics/faculty/bios/HaneyDocs/page-58978134.html [Accessed 24 October 2013].

Alden, L., (2005). Statistics can be misleading [online]. Available from: http://www.econoclass.com/misleadingstats.html [Accessed 25 October 2013].

Author unknown, . Discover How to Make Money Online and Incomes [online]. Available from: http://www.calvin.edu/academic/economics/faculty/bios/HaneyDocs/page-59426962.html [Accessed 24 October 2013].

Freebern, N., (2001). Pictograph [online]. Oswego City School District. Available from: http://www.studyzone.org/testprep/math4/d/pictogra4l.cfm [Accessed 31 October 2013].

Author unknown, . Chapter 6: Pretty Pictures [online]. Available from: http://www.calvin.edu/academic/economics/faculty/bios/HaneyDocs/page-58978264.html [Accessed 24 October 2013].

Author unknown, . A Wrong Way and a Right Way [online]. Available from: http://www.calvin.edu/academic/economics/faculty/bios/HaneyDocs/page-59480389.html [Accessed 25 October 2013].

Author unknown, . Misleading Graph #2 [online]. Available from: http://www.calvin.edu/academic/economics/faculty/bios/HaneyDocs/page-59448244.html [Accessed 25 October 2013].



                In conclusion, statistics can be either reliable or unreliable. Unreliable statistics can be intentionally misleading or unintentionally misleading depending on the aim of the researchers. Moreover, statistics can be used to trick people for personal benefits. In addition misinterpretation of statistics can directly lead people to fall into traps which have been statistically set up. Therefore, it is essential to not make generalizations without enough experience and knowledge. Instead, it is best to deduce the conclusion from the data with caution and careful thought.