Statistics- Facts Or Fallacies Essay, Research Paper
The most common signifier of grounds is statistics. Statisticss are a favourite grounds of many authors and talkers. They provide existent Numberss in support of thoughts and decisions. If you can demo that 75 % of high schools seniors can non happen Washington State on a map of North America, so it is strong grounds for your contention that high school seniors are non being taught the geographics of the United States. Such grounds is non merely hard to rebut, it & # 8217 ; s frequently accepted as the concluding word in what & # 8217 ; s true or non true.
Statisticss are a premier beginning of cogent evidence that what you say is true. Statisticss are based on surveies: a hunt for possible connexions between disparate facts that however have a connexion. If you remember your math categories, you will remember the construct of sets and subsets. Statisticss are, in big step, concerned with that construct. They are fundamentally stating you the proportion a subset represents in a set. To clear up this thought, look at political polls. Candidate A receives 46 % blessing, Candidate B receives 43 % blessing. Therefore, the subset & # 8220 ; responses prefering Candidate A & # 8221 ; is 46 % of the whole set, & # 8220 ; People asked about Candidates A and B. & # 8221 ;
Another illustration, from existent life. William Chadwick, with his helper William Farr, during the great cholera pestilence in London in 1831, drew together factors on who was acquiring the disease and where they were acquiring it in London. They were looking for some common factor that would take to what was the beginning of the disease. Their statistics led them to the decision that the contaminated Waterss of the Thames River was the beginning, and at that place was a peculiar pump that supplied the H2O to certain neigh borhoods that was a premier beginning of infection. With these informations they were able to do recommendations which did much to cut down the incidence of cholera in London.
Statisticss besides use samples to obtain consequences, instead than making existent & # 8220 ; caput counts & # 8221 ; . Neilson evaluations on how many of what sort of people watch a peculiar Television plan is non determined by the Neilson company inquiring all 300 million people in the Uni ted States what they are watching every few proceedingss. What they use is a sample of the population ( called the Neilson households ) that, demographically, represent the 300 million people. Neilson selects these households really carefully since each one represen ts the screening wonts and desires of some 60,000 people. However the statistics generated by the Neilson measurings are used to do scheduling determinations and put advertisement rates and budgets, things that represent one million millions of dollars. Thus the se lection of the sample, whether Neilson & # 8217 ; s or incidence of AIDS in the US population, is of paramount importance in the cogency of the statistics therefore generated.
The above is, of class, a simplistic position of an highly complicated subject. It is, however, the kernel of statistics.
Statisticss are priceless as grounds in support of decisions. If you can either happen or bring forth statistics that show the truth of your decisions, there are few that would rebut your thoughts.
There are, of class, jobs with utilizing statistics as grounds. Let me remind you of a celebrated expression: & # 8220 ; There are three ways to non state the truth: prevarications, damned prevarications, and statistics. & # 8221 ; What you must make is inquire yourself some inquiries: who did the survey that came up with the statistics, what precisely are the statistics measurement, who was asked, how were they asked, and compared with what? If one believes in the truth of statistics ( and there are many such ) , so how does one explain that the same Presidential campaigner can be 20 points in front and 5 points behind his opposition in the polls at the same clip? After all, both polls are & # 8220 ; statistics & # 8221 ; . What you must be examine, if you wish to utilize statistics as grounds, are the above inquiries.
Who Did the Survey
Let us analyze first & # 8220 ; who did the study. & # 8221 ; We live in a universe of statistics: you can happen Numberss in support of merely about any thought. The job arises when you find statistics that support every manner of sing an thought. You can happen statistics that show coffin nails are slayers and that they have no consequence on anyone & # 8217 ; s wellness. You can happen statistics that say you should cut down on the ingestion of dairy merchandises and that dairy merchandises are good for you. You can happen statistics that prove that so ft drinks will give you malignant neoplastic disease and that they have no consequence on anything but your thirst ( or even that they make you thirstier ) . Every one of these sets of statistics is perfectly true.
The phrase & # 8220 ; Numberss don & # 8217 ; t lie & # 8221 ; is true ; what you need to analyze is who is printing the Numberss, and what are they seeking to turn out with them. Are the statistics provided by the American Cancer Society or the American Tobacco Institute? Are they provided by the American Medical Association or the American Dairy Association? Are they provided by the Cancer Institute or the United States Food and Drug Administration? ( Did the latter give you pause? It should. Both are reputable. Yet both have differing sentiments based on statistics. )
Every point of position uses statistics to back up their thoughts. It & # 8217 ; s your occupation to analyze all statistics back uping all points of position, to get at your ain decisions based on all of them. If you can & # 8217 ; t arrive at a decision, make your ain survey. An easier class, of course, is to happen out what all possible sides have to state and what other grounds they have in support of their statistics.
Once you have determined whether or non there is bias involved in the statistics ( delight remember that subjectiveness is ineluctable ) , so it is clip to travel on to the following inquiry: what are the statistics mensurating?
What are the Statisticss Measuring
When inquiring yourself, & # 8220 ; what are the statistics measurement, & # 8221 ; bear in head the old proverb about mensurating apples and oranges. Most people will state that you can & # 8217 ; t compare apples and oranges. This is both true and false. It depends on WHAT YOU ARE MEASURI NG. Color? No. Texture? No. Overall visual aspect? No. Acidity? Yes. Sugar content? Yes. Vitamin, mineral, saccharide, or flesh out content? Yes.
As you can see, it is possible to compare apples and oranges, if you know what you are mensurating. Your occupation, in utilizing statistics as grounds, is to find what precisely is being measured, and non merely spout Numberss that seem to use to your subject. If your subject is & # 8220 ; Nutritional Value of Oranges, & # 8221 ; statistics turn outing that apples are nil like oranges may be mensurating the incorrect things.
Who was Asked?
Once you & # 8217 ; ve determined what the statistics are mensurating, you next need to happen out how the research was done. Many surveies, the consequences of which are disseminated utilizing statistics, are done by inquiring people their sentiments or what they do or believe or experience or. . .. Such surveies include political, sociological, consumer behaviour, media audience, and other countries which are based on single people & # 8217 ; s thoughts, sentiments and/or attitudes.
Such countries are frequently referred to as & # 8220 ; soft scientific disciplines & # 8221 ; , as opposed to & # 8220 ; difficult scientific disciplines & # 8221 ; that do research designed to minimise every bit much as possible the human factor in the grounds and decisions. The & # 8220 ; human factor & # 8221 ; is, of course, impossible to extinguish wholly every bit long as worlds are involved, but the surveies, to be & # 8220 ; scientific, & # 8221 ; must be quotable and prognostic in nature. That is, one time a survey has been done, tantamount consequences must look when the survey is done once more by other research workers who have no connexion with the original research workers, and the consequences should let research workers to state what will go on next.
Let us state that scientific statistics show meteors fall during a specific period ( say, August ) at an mean rate ( say, 60 per hr ) . This survey is repeated several old ages during August and the rate stays the same. Thus the survey is quotable. From those statistics it is possible to foretell that in future old ages the mean rate of hiting stars in August will go on to be 60 per hr. In this instance, & # 8220 ; who is being asked & # 8221 ; are the impersonal forces of nature.
It is the soft scientific disciplines that most frequently, deliberately or accidentally, abuse or misapply statistics. The surveies are frequently non quotable and normally non prognostic. The ground for this is that people and what they say or do are the bases of T he statistics. It seems self-evident that people will perversely decline to state or make the same thing twice running, or allow anyone foretell what they will make. In fact, many people consider themselves insulted when called predictable, and anything from the conditions to the clip of twenty-four hours to who & # 8217 ; s inquiring the inquiry can alter what they will state or make about something.
What does this mean to you as you examine the statistics you plan on utilizing as grounds? First, attempt to find whether the statistics are difficult or soft scientific discipline based. The simplest manner to make this is merely happen out if people or nature is being studied. If nature it & # 8217 ; s difficult scientific discipline, if people it & # 8217 ; s soft.
Second, if the statistics are difficult scientific discipline, cheque to see what consequences other research workers who have repeated the survey obtained. If the 2nd survey has consequences that vary widely from the first, happen a 3rd and/or 4th and utilize the consequences that are consistent overall.
Of class, difficult scientific discipline statistics frequently require that you examine who was asked. Check the sample: if the statistics say that 30 % of the US population has AIDS, what was the sample? The full population of the US? The population of New York or San Francisco? The population of Otumwa, Iowa? Or a choice of towns and metropoliss, rural, urban and suburban, in all parts of the state? Statisticss on the incidence of colza in the US vary wildly depending on whether the survey asks jurisprudence enforcement or R ape guidance centres ( one set is based on the figure of reported colzas, the other on the figure of adult females necessitating reding whether or non they reported the colza to jurisprudence enforcement ) . Both illustrations above appear to be difficult scientific discipline, since they are based on & # 8220 ; difficult & # 8221 ; facts, but however must be examined for who was asked.
Soft scientific discipline statistics are even more slippy than difficult scientific discipline statistics. First, there are few difficult, quotable, non-subjective facts on which to establish the statistics. If you wish to demo how people react to violence, how do you specify force? And how make the people in your survey define force ( a victim of a mugging may specify force as acquiring within five pess of him, while a mugger may specify it as anything that happens that causes him physical harm ( what he does to others is merely high liquors ) ) .
Besides bear in head that any survey that uses human topics is about impossible to carry on under research lab conditions, in which all factors that could consequence the result of the experiment are controlled, including the variable under survey. For a genuinely statistically valid survey demoing the effects of telecasting force on kids, the kids would hold to isolated from all other factors that could hold an influence. These other factors would include contact with other human existences, with other looks of force ( people, reading, wireless, films, newspapers, picture games, etc. ) . This would evidently work to the societal and developmental hurt of the kids.
As a affair of fact, a recent contention arouse over utilizing medical informations collected by the Nazis in the concentration cantonments. These informations were collected with perfectly no respect for the fact that the trial topics were human existences ; they were treated much worse than any laboratory animate being in the universe today. Ethical and moral considerations aside, the informations are viewed as valuable. However, there are people who believe that the ethical and moral considerations are paramount, and that the informations, no affair how valuable, should be destroyed because of the manner they were gathered.
In add-on to the fact that any survey affecting worlds must take into history human and humanist considerations, you should ne’er undervalue the contrariness of a human being. In analyzing comedy one of the first things I learned was ne’er tell the audience I was traveling to be amusing. The minute a comedian says to an audience, & # 8220 ; You & # 8217 ; re truly traveling to happen this good story, & # 8221 ; the same audience that minutes before was falling out of their chairs express joying will turn cold and soundless, with an & # 8220 ; Oh, yeah? Show me & # 8221 ; attitude.
In the same vena, a truism in advertisement is that 50 per centum of advertisement plants ; the job is no 1 can calculate out which 50 per centum. The ground is that no 1 can truly calculate out what will act upon people to purchase merchandises.
To seek to understand & # 8220 ; soft & # 8221 ; statistics, allow & # 8217 ; s take a expression at advertisement research and consumer behaviour, both of which are subsets of socio- and psychological research. In peculiar, we & # 8217 ; ll look at some basic maxims of consumer research that apply to any soft statistics.
First is the realisation that all people are different. No two people, non even indistinguishable twins, are precisely the same background and upbringing, have had the same conversations in the same words, have read the same books or magazines or newspapers at precisely the same clip, or done anything the same as anyone else. This fact is exactly the antonym of what is necessary to statistics & # 8212 ; that there are similarities that give significance to the variables.
There are, of class, some factors that many people have in common with other people, and upon them statistics depend. These factors can include the society in which they live, their societal category, whether they are urban, suburban or rural ; their relationships & # 8212 ; most people have had a female parent and male parent, possibly siblings, friends of the same or opposite sex ; and their involvements: athleticss, telecasting, reading scientific discipline fiction or enigmas or love affairs. Of class, non everybody fits into all classs. Again, all people are different, but they do hold some things in common.
What the above means is that no statistic has any application to an person, but can hold an application to the group. However, the statistics are determined on the footing of analyzing persons in the group, non analyzing the group. Now recall the jobs with persons. First, persons change, non merely from twelvemonth to twelvemonth but from minute to minute.
Second, persons are inconsistent. What they like today they may detest the following. You may love spaghetti, but eat it five yearss in a row, and you may happen the idea of eating it once more sickening.
Third, persons frequently don & # 8217 ; t know what they want, and even if they do, they don & # 8217 ; t know or can & # 8217 ; t state you why.
Then there are a few jobs involved in appraising persons togather the information to explicate the statistics. First, people frequently can & # 8217 ; t retrieve information about themselves and therefore the background can be uncomplete. If you don & # 8217 ; t believe this, remember precisely when you got your last tetanus supporter shooting, or the class you got in first-year English in high school.
Second, there is a prestige prejudice. Answers a individual gives involve the individual personally & # 8212 ; his or her pride, self-esteem and self-image are involved. Therefore people will frequently give an reply that will rise their image. Harmonizing to Television sing diar Internet Explorers, cipher tickers professional adult females & # 8217 ; s wrestle, but Masterpiece Theatre has a 50 evaluation. In some categories a few old ages ago I ran a study that, as a portion of the background, asked & # 8220 ; How many hours do you watch telecasting during an mean week. & # 8221 ; The mean reply was seven hours per hebdomad ( delight remember that the national norm is seven hours per twenty-four hours ) . Granted, college pupils do non normally have a great trade of clip to give to watching Television, but the categories in which I gave this study were advertisement and mass media unfavorable judgment, both of which require watching telecasting. What & # 8217 ; s more, for people who avowed small involvement in telecasting, these same pupils had a close encyclopaedic cognition of inside informations about plans and/or commercials that were discussed, in many instances equaling my ain ( I watch telecasting an norm of eight hours per twenty-four hours ) . It was clear that the responses on the study bore small relationship to world. Nonetheless, I was non surprised at the responses. Television watching traditionally has a prestigiousness job, and prestigiousness bias clearly influenced how people answered the inquiry.
Third, people lie. That may look a spot blunt, but there is no ground to glaze. Peoples non merely stretch the truth, story or misspeak themselves. They lie. Ask them a inquiry and, merely for the snake pit of it they may lie. They may lie because they find the truth uncomfortable or awkward, or because they merely want to sleep together up your consequences. With lying a practical societal necessity ( make you truly state your best friend that his or her breath could strike hard a turkey vulture off a honey waggon? ) , the fact the people lie when reacting to surveies should come as no surprise.
Finally, many surveies non merely seek to happen out what people do, but why they do it. Here the job lies in respondents & # 8217 ; inability to joint or explicate their true feelings and motives. Many people do things because it & # 8220 ; feels & # 8221 ; like the thing to make, but they can non explicate what that feeling is or how it arose. They will make the best they can, but since so many such feelings are subconscious and/or based on a priori premises, they have ne’er been examined and put into words.
How Were They Asked?
It is non merely the respondents but the inquirers that contribute their ain bias to the assemblage of facts.
Two things that are used in studies and statistical surveies are inquiries and replies. First, allow & # 8217 ; s analyze the inquiries.
Research workers by and large have an thought what their research is looking for. They therefore formulate inquiries that will light their research, either pro or con. Prejudice can crawl in when a research worker unconsciously words inquiries in such a manner that the replies back up his or her contention or sentiment. Assorted inquiries of this type are taking inquiries, laden inquiries, and double-barreled inquiries.
Leading inquiries are those that tell the respondant how to reply. Attorneys sometimes use them. For illustration, & # 8220 ; Is it non true that on the dark of the 27th you were drunk? & # 8221 ; Such a inquiry leads the respondent to state yes. Asking alternatively, & # 8220 ; Were you drunk on the dark of the 27th? & # 8221 ; does non state the informant how to react.
Loaded inquiries are those that, no affair how they are answered, the respondent loses. & # 8220 ; Are you still crushing your married woman? & # 8221 ; and & # 8220 ; Are you still rip offing on your income revenue enhancement? & # 8221 ; are illustrations. A laden inquiry appears to inquire for a yes or no reply, yet the existent reply may be neither yes nor no.
Double-barreled inquiries are those that ask for more than one piece of information in the same inquiry. For illustration, & # 8220 ; Do you travel up or downtown in the afternoon? & # 8221 ; is double-barreled.
Another point to be considered is how the inquiries were worded. It is easy, and frequently subconscious, for the inquirer to word the inquiries in such a manner as to take to respondent to answer in a certain manner. For illustration, a study on whaling could inquire, & # 8220 ; Should the lone three states in the universe that do so, continue to butcher to extinction the helpless, harmless intelligent giants of the deep? & # 8221 ; I surmise that few people would react with a yes.
It is the replies that sometimes cause trouble for a research worker. The jobs lie non merely in how the respondents answer, but in how the research worker responds to the reply. Sometimes the response is non what the research worker wants or needs and/or contradicts outlooks. He or she must so account for the anomalousness. He or she may revamp the original construct or theory, revamp the survey, or even ignore the information. The research worker may fall quarry to selective perceptual experience ( seeing merely what you want to see ) or cognitive disagreement ( apologizing off anything that doesn & # 8217 ; t fit into your prepossessions ) . In add-on, how the research worker interprets the words in the inquiries may be at odds with how the respondents interpreted the words. For illustration, in a recent study on the incident of colza on college campuses, the inquiries used words such as unwelcome sexual progress ; the research worker interpreted unwelcome sexual progress as colza, while the respondents could good hold been mentioning to a rummy at a saloon doing a base on balls, something that most people would accept as disgusting, but non colza.
The order of the inquiries can besides be a job. Often, the inquiries can take a respondent to reply in a certain manner because he or she has answered all the old inquiries in the same manner. In gross revenues, it & # 8217 ; s a common technique, that can take a respondent through a series of yes replies, from & # 8220 ; it & # 8217 ; s a nice twenty-four hours, & # 8221 ; to & # 8220 ; mark here. & # 8221 ;
Therefore & # 8220 ; How were they asked? & # 8221 ; requires an scrutiny of the original survey in order to see if the research worker may hold made an mistake in oppugning and in understanding the replies.
Compared with What?
Finally, you need to analyze statistics to find what are the comparings being drawn and are they relevant and valid. For illustration, state your subject is gun control. You could happen statistics on slaying rates with pistols per capita in New York City, London and Tokyo. Such statistics would demo much higher rates in New York than the other two metropoliss. It would hence look that gun control is a good thought since guns are controlled in London and Tokyo. However, such statistics must be fishy, non because they are incorrect ( more people are so murdered with pistols in New York City than in London or Tokyo ) , but because they don & # 8217 ; t state the whole narrative.
For case, New York has an highly rigorous arms control jurisprudence ( the Sullivan Act ) . Since this is the instance, what happens to the statement that control Torahs work? There must be something else act uponing the slaying rate.
What about the civilization? The United States is unlike any other state on Earth. Its society has a tradition of independency and autonomy, where if you have a job it is normal for you to take attention of it yourself, even if you can & # 8217 ; t. It is besides a state that used to be called & # 8220 ; the melting-pot & # 8221 ; but is now known as the & # 8220 ; mosaic & # 8221 ; , with New York City a hodgepodge of frequently conflicting civilizations, linguistic communications, imposts and attitudes. Add in the traditions of the old West and & # 8220 ; gunslinging & # 8221 ; becomes an seemingly feasible option to work out jobs. Japan, on the other manus, is an highly homogeneous and traditional civilization, with small in the manner of open category or cultural struggle. England is besides really traditional with far less cultural struggle ( any state that feels no necessity to build up their constabulary does non hold a tradition of single usage of force to work out jobs ) . However, now as England is going more culturally and ethnically diverse, there is a lifting incidence of force and usage of guns.
From the above it is clear that any statistics on slaying rates says nil about the efficaciousness of gun control Torahs, but instead about the cultural and/or social factors that make such Torahs ineffective. If you wish statistics to function as grounds for a gun control jurisprudence, find something else.
For the above grounds you must seek for other grounds to back up whatever statistics you use as support, if merely to demo that the statistics really apply.
Do non, nevertheless, take all the jobs outlined supra as a disapprobation of statistics as grounds. Statisticss are first-class grounds, and frequently the easiest and most concise manner to show grounds. I simply wish you to be cognizant you must analyze them for relevancy, cogency and authorization or they can make you more injury than good in turn outing your point.