the truth about public opinion polls

                                          The Truth about Public Opinion Polls

            There is an old wives tale about the best way to boil frogs. The steps are relatively simple. After catching a frog, it is placed into a pot of tepid water.  The water doesn’t bother the frog and he is comfortable enough in the pot.  Next, slowly turn up the heat in the pot. The frog, acclimating to the warmer water stays comfortable enough. Finally, turn up the heat slowly.  The frog, not noticing the slow increase in temperature is boiled before it knows any better.  This appears to be the same routine that has accrued with the United States and public opinion polls.  Polling in the U.S. has become so prolific that it is almost impossible to pick up any newspaper, article or watch any news program without being bombarded with the ever-popular opinion poll.  Generally suffocating the public with “unimportant, alleged opinions”, (Vatz, 2006) polls have become an out of control reality of everyday America.

            Polling in the United States dates back to the local straw polls in the early 1800’s. While trying to estimate who would win the upcoming presidential campaign, many cities would hold local opinion polls.  However, these “straw polls” were neither systematic nor scientific and generally just tallied the opinions of whoever chose to take part. (Encarta, 2009) with the success of the first several presidential races being predicted correctly however, people took the polls as representing a good predication of future events.  It was not until the fatal prediction by the Literary Digest in 1936 that Alf Landon would win the presidency over Theodore Roosevelt that polling techniques began to be taken seriously.  The Literary Digest poll determined that Alf Landon would win by a majority.  When this failed to happen, not only did the magazine go under, but polling agencies wanted to know why.  They uncovered the error of the based nature of the poll, which was only representing the wealthy republican class, which reflected the subscribers of the publication. (Encarta, 2009)   The respectability of opinion polls started to gain influence and in the beginning was based on communication research. (Herbst, 1993) With a more classical view, public opinion took into consideration theory, history, or the generation of public opinion through political philosophers. They warned against the misuse of using polls politically, cautioning politicians against becoming, “flatterers instead of legislators; the instruments, not the guides, of the people.”(Knull, 2002)  However, this was quickly replaced by a more social science approach that concluded public opinion was “simply an aggregation of public opinions, as captured by a survey.” (Herbst, 1993) It was George Gallup’s scientifically-based methods that caught the attention of the public opinion research world when his small but very accurate prediction of Roosevelt’s landslide win was correct.  Although his methods were crude compared to today’s polling advances, his theories stuck, and more classical approaches disappeared. Regardless, this was the end of straw polling, and the beginning of a more educational and scientific approach to public opinion polls.

            An opinion poll, simply put, is the “measurement of specific issues through interviews with a representative sample of the group whose views are to be described.” (Encarta, 2009) Commonly, poll topics are used for political intentions in regards to party support, government policies, and views on major public issues.  However, in today’s poll-happy atmosphere, public opinion polls are regularly used to drive media, marketing, and even the making of state policies.  Because public opinion is so easily molded by circumstance, polling issues can have a great impact on the collective conscience of society.  Although some issues, like race, religion and geographic location are relatively permanent they can have a strong impact on the way people think about certain subjects. Other more temporary issues, like media, public relation campaigns, and current events can also change the collective opinion of the public. (Encarta, 2009)

            This potential power to control the thinking of large numbers of people by using polling techniques is an alluring tool for businesses, politicians, media, government and even academic research. (Encarta, 2009) Conducting polls not only gives these organizations a tool by which to measure preferences, but also an outlet in which to persuade, although generally unknown to the public. Businesses use polls to discover what makes a product more appealing in order to increase sales.  Politicians use polling to uncover where the majority of voters’ stand on a certain issue in order to prepare a successful campaign.  Often they will avoid topics where their opinion would not be popular, and spend a lot of time on topics that are important to voters. In addition, the media is addicted to using public opinion to drive stories and get attention.  Often they misuse information in order to deliver a shocking headline. For instance, in one headline on MSNBC that “Iraq replaces jobs as most important in American priority.”(Vatz, 2006) What the reporter neglected to mention was that the level of concern had risen from 39% to 40%, and that job concern had dropped from 46% to 35%.  Had war concern actually increased, or did a stronger employment situation just clam fears about job concerns. (Vatz, 2006) The manipulations of polls to drive stories have a tendency to mislead and cause the problems they are errantly reporting. Although polling can be helpful, it is generally the misuse of polling that is running rampant and battling for control of the American conscience.

            This dilemma is not unrecognized, even by the most well-known and respected of the polling companies.  Gallup, still one of the most popular polling agencies, tries to help the organizations, such as the media, to use polls correctly.  As a general rule, Gallup strongly suggests that using only polls that are conducted in a scientific manner be used at all.  The major difference is that non-scientific polls are taken by people volunteering, for example, at a mall or internet polls, while scientific polls are taken from a premeditated cross section of the United States.  A non-scientific poll, being biased, will not project the opinions of Americans as a whole, and only gives insight into a small sub-group of the population. (Witt, 2009)  For instance, if the poll was taken at a mall, only people in that specific geographic location, who go to the mall, and who volunteered to poll, will be included in the polls outcome.  Gallup also provides guidelines to help decipher between a reliable versus unreliable poll.

            One of the most important aspects of the poll is the question of who took the poll. This is important largely because of the importance of finding out vital information like, who paid for the poll to be taken, how many people were interviewed, how and where they were chosen, other factors that can skew results and exactly what questions were asked.(Witt, 2009)  These questions, and others, can only be answered by the polling company.  And although some questions seem a bit simplistic, their answers are vitally important.  As mentioned before, the methods use to measure public opinion can vary greatly.  Some poll companies are trusted as good sources of information because of their ability to follow a certain set of standards that allow for good answers to the above questions.  For instance, just finding out who paid for the poll can reveal many things about the trustworthiness of that particular poll. “Polls are not conducted for the good of the world. They are conducted for a reason – either to gain helpful information or to advance a particular case.” (Witt, 2009)   Although the main reasons to conduct polls, whether by the politician, the businessman, or the marketer, are all good, the motive for the poll must reinforce the increase its validity.  An example of this can easily be drawn from the political realm.  Polls conducted with slanted questions, or with the hidden agenda of mudslinging an opponent will not have accurate results of public opinion.

            Also of great importance is the number of people included in the poll. By using correct principals, a polling company can select 1000 individuals and using the statistics from that control group accurately predicts the opinion of 210 million people. In a scientific poll the pollster using statistical techniques to choose an appropriate cross section of people that will represent the whole. Thus the number of people and the people chosen to cooperate in the poll are all important factors. Another good illustration of this is a poll that using telephone surveys to answer questions about the difficulties of being homeless, the poll would be inaccurate.  People in homes with telephones, are not experiencing homelessness. (Witt, 2009)

            Moreover, in addition to the many factors that are important in tallying a good representation of real public opinion, errors that might skew the poll are of utmost importance.  One such error could lie in the tricky territory of sampling errors.  The fact of the matter is that there is a slight percentage for error in every poll. Pollers take this into consideration by allowing what is called a sampling error.  Using statistical methods, pollers can evaluate closely the outcome of a poll using percentages.  If, for instance, in a political race, the candidate’s race is very close, reporting that one candidate is clearly ahead of the other would be false, taking into consideration the six or so percent taken into account as a sampling error. But sampling errors are minimal compared to the complexity of errors that arise from question phrasing and question order. (Witt, 2009)  And most importantly, exactly what questions were asked can contribute greatly to the outcome of the poll. Questions that are unfairly biased, or that give an unbalanced set of choices are good examples of bad questioning practices.  Special-interest groups and the media are specifically at risk for errors of this kind when seeking public opinions on current-issues.  This can especially be seen in the media.  “Predetermined categories make it appear that respondents in large numbers feel the same way.”(Vatz, 2006)

But polls are not open-ended and generally oversimplify questions to complex issues and supply one or two options that don’t totally reflect the feeling of the public. An illustration of this is in a poll asking whether the use of atomic bombs 60 years ago was avoidable. It does not define avoidable. It does not mention how many American lives were saved. The article then mentions that “Two-thirds of Americans say the use of atomic bombs was unavoidable, but that only 20 percent of Japanese felt that way and three fourths said it was not necessary.”(Vatz, 2006)

            With so much room for error it is easy to see why there is so much controversy behind the benefits of public opinion polls. However, if no one cares about the importance of good polls, “cheap, dirty polls will surely drive out the good ones.” (Taylor, 2004)  Just as there are methodologies and techniques for good polling, so there are will bad polling. “To much of the media, a poll is a poll is a poll.”(Taylor, 2004) Many bad polls are obvious, although that doesn’t stop the media from using them.  However, the factors that make up a bad poll are easily recognizable.  For instance, bad polls survey less people, which is cheaper. Bad polls usually return a surprising result, use a one-day instant poll method, or use the 900-number straw poll technique. Also, bad polls oversimplify issues, use focus groups, or have manipulated word order in the questions.  With so much at stake, bad polling should be reprimanded by not being reported, thus ending a vicious cycle.

            Moreover, with Americans sentiments that “too much polling seems like overkill.”(Robertson, 2002)  At the same time however, skeptism about polling doesn’t seem to have affected America’s appetite for percentages. “In 2007, 71 percent of all media stories were driven by poll numbers.”(Knull, 2004) And even more telling is the fact that in the majority of these cases, “the poll itself became the news.”(Knull, 2004)

            This has become detrimental in the fact that bad polling that gets reported then begins to skew and manipulate the thinking of the public, sometimes even outweighing and undermining the legislative process.  This can be seen clearly in the disturbing manner in which the Supreme Court sometimes makes it rulings.  In one case, in which a statute permitting the execution of a mentally retarded murderer was being deliberated, the case was overturned when the Supreme Court used public opinion as the basis for their decision. (Knull, 2004)  One outraged reporter speaking out on the subject argued that

“Pollsters must have amended the 19th century dictum that the Supreme Court follows the election, to the Supreme Court follows the ups and downs of the pollsters’ claims of public opinion.”(Knull, 2004)

This outrageous phenomenon is disturbingly branching out in other facets of legislature as well.  As is easily seen in the realm of morality policies that deal with issues like drugs, alcohol and abortion.  These are highly charged topics that are not terribly complex and that most people have strong opinions about. And the fact of the matter is that people’s opinions influence public policy. “Public opinion has been demonstrated to match public policy in the United States.”(Norrander, 1999)  And because public opinion drives the election process, which is intimately tied to the policy making process, the accuracy of trustworthy public opinion polls could not be greater.

            This sheds an interesting light on the  Lewinsky Scandal during the presidential term served by Bill Clinton.  Amazingly, after the initial story of the scandal was released the public,measured by polls, was said to have not reacted to the shocking news with much change of opinion.  Although the scandal grew from it’s beginnings as a sexual scandal into an issue of perjury, still the polls suggested Clinton’s popularity had not been much affected (Bennett, 2002).  Impeachment seemed inevitable and the Senate tried to convict Clinton and failed. “Clinton allegedly exclaimed, ‘Thanks god for public opinion.”(Bennett, 2002)  Students of public opinion have tried to prove that this occurrence was not just fluke, but more a result of the fact that the public simply doesn’t pay attention to public affairs.  “An inattentive public is therefore a hindrance and a help to the president.”(Baum, 2004)  In any case it is obvious that public opinion polls have a large impact on the shaping of the country, whether it is acknowledged or not.

In summary, although any polling agency would claim that they only measure public opinion, in truth they play a large part in shaping public opinions.  From affecting elections to persuading decisions in the Supreme Court polling makes an impact. President Bush stated in a press conference that “ I don’t think you can make good, sound decisions based on polls.”(Vatz, 2006)  Although public opinion shouldn’t be ignored completely, it is obvious that the measurement of public opinion is flawed.  That

“imperfection is compounded by instability, difficult to measure intensity, woefully inadequate information and short term policy preferences.” (Vatz, 2006) Although we are better informed with them, poll readers should keep in mind the different methods, reliability and motives of pollsters.


Baum, M. A. (2004). How Public Opinion Constrains the Use of Force: The Case of Operation Restore Hope. Presidential Studies Quarterly, 34(2), 187+. Retrieved March 19, 2009, from Questia database:

Bennett, S. E. (2002). Another Lesson about Public Opinion during the Clinton-Lewinsky Scandal. Presidential Studies Quarterly, 32(2), 276+. Retrieved March 19, 2009, from Questia database:

Herbst, S. (1993). History, Philosophy, and Public Opinion Research. Journal of Communication, 43(4), 140-145. Retrieved March 19, 2009, from Questia database:

Knull, M. (2002, October). Slanted Polls Yield Biased News. World and I, 17, 60. Retrieved March 19, 2009, from Questia database:

Norrander, B., & Wilcox, C. (1999). Public Opinion and Policymaking in the States: The Case of Post-Roe Abortion Policy. Policy Studies Journal, 27(4), 707. Retrieved March 19, 2009, from Questia database:

“Public Opinion,” Microsoft® Encarta® Online Encyclopedia 2008 © 1997-2008 Microsoft Corporation.

Robertson, L. (2001, January). Polled Enough for Ya?. American Journalism Review, 23, 24.             Retrieved March 19, 2009, from Questia database:

Taylor, H. (1992, October 19). Can Bad Polls Drive out Good?. National Review, 44, 48. Retrieved March 19, 2009, from Questia database:

Vatz, R. E. (2006, May). Of What Value Are Public Opinion Polls?. USA Today (Society for the Advancement of Education), 134, 62+. Retrieved March 19, 2009, from Questia database:

Witt, Evans. “20 Questions A journalist Should Ask About Poll Results.” Retrieved March 18, 2009.


"Looking for a Similar Assignment? Order now and Get a Discount!

"Looking for a Similar Assignment? Order now and Get a Discount!

Posted in Uncategorized