opinion poll

11
Opinion poll An opinion poll, sometimes simply referred to as a poll, is a survey of public opinion from a particular sample. Opinion polls are usually designed to represent the opin- ions of a population by conducting a series of ques- tions and then extrapolating generalities in ratio or within confidence intervals. 1 History The first known example of an opinion poll was a lo- cal straw poll conducted by The Harrisburg Pennsylva- nian in 1824, showing Andrew Jackson leading John Quincy Adams by 335 votes to 169 in the contest for the United States Presidency. Since Jackson won the popular vote in that state and the whole country, such straw votes gradually became more popular, but they re- mained local, usually city-wide phenomena. In 1916, the Literary Digest embarked on a national survey (partly as a circulation-raising exercise) and correctly predicted Woodrow Wilson's election as president. Mailing out millions of postcards and simply counting the returns, the Digest correctly predicted the victories of Warren Hard- ing in 1920, Calvin Coolidge in 1924, Herbert Hoover in 1928, and Franklin Roosevelt in 1932. Then, in 1936, its 2.3 million “voters” constituted a huge sample; however, they were generally more afflu- ent Americans who tended to have Republican sympa- thies. The Literary Digest was ignorant of this new bias. The week before election day, it reported that Alf Lan- don was far more popular than Roosevelt. At the same time, George Gallup conducted a far smaller, but more scientifically based survey, in which he polled a demo- graphically representative sample. Gallup correctly pre- dicted Roosevelt’s landslide victory. The Literary Digest soon went out of business, while polling started to take off. Elmo Roper was another American pioneer in political forecasting using scientific polls. [1] He predicted the re- election of President Franklin D. Roosevelt three times, in 1936, 1940, and 1944. Louis Harris had been in the field of public opinion since 1947 when he joined the Elmo Roper firm, then later became partner. In September 1938 Jean Stoetzel, after having met Gallup, created IFOP, the Institut Français d'Opinion Publique, as the first European survey institute in Paris and started political polls in summer 1939 with the ques- tion "Why die for Danzig?", looking for popular support or dissent with this question asked by appeasement politi- cian and future collaborationist Marcel Déat. Gallup launched a subsidiary in the United Kingdom that, almost alone, correctly predicted Labour’s victory in the 1945 general election, unlike virtually all other commen- tators, who expected a victory for the Conservative Party, led by Winston Churchill. The Allied occupation powers helped to create survey in- stitutes in all of the Western occupation zones of Ger- many in 1947 and 1948 to better steer denazification. By the 1950s, various types of polling had spread to most democracies. 2 Sample and polling methods Opinion polls for many years were maintained through telecommunications or in person-to-person contact. Methods and techniques vary, though they are widely ac- cepted in most areas. Verbal, ballot, and processed types can be conducted efficiently, contrasted with other types of surveys, systematics, and complicated matrices beyond previous orthodox procedures. Opinion polling developed into popular applications through popular thought, although response rates for some surveys declined. Also, the following has also led to differentiating results: [1] Some polling organizations, such as Angus Reid Public Opinion, YouGov and Zogby use Internet surveys, where a sample is drawn from a large panel of volunteers, and the results are weighted to reflect the demographics of the population of interest. In con- trast, popular web polls draw on whoever wishes to par- ticipate, rather than a scientific sample of the population, and are therefore not generally considered professional. Recently, statistical learning methods have been proposed in order to exploit Social Media content (such as posts on the micro-blogging platform of Twitter) for modelling and predicting voting intention polls. [2][3] Polls can be used in the public relation field as well. In the early 1920s Public Relation experts described their work as a two-way street. Their job would be to present the misinterpreted interests of large institutions to public. They would also gauge the typically ignored interests of the public through polls. 1

Upload: betafilip6910

Post on 03-Feb-2016

15 views

Category:

Documents


0 download

DESCRIPTION

opinion poll of my class

TRANSCRIPT

Page 1: Opinion Poll

Opinion poll

An opinion poll, sometimes simply referred to as a poll,is a survey of public opinion from a particular sample.Opinion polls are usually designed to represent the opin-ions of a population by conducting a series of ques-tions and then extrapolating generalities in ratio or withinconfidence intervals.

1 History

The first known example of an opinion poll was a lo-cal straw poll conducted by The Harrisburg Pennsylva-nian in 1824, showing Andrew Jackson leading JohnQuincy Adams by 335 votes to 169 in the contest forthe United States Presidency. Since Jackson won thepopular vote in that state and the whole country, suchstraw votes gradually became more popular, but they re-mained local, usually city-wide phenomena. In 1916,the Literary Digest embarked on a national survey (partlyas a circulation-raising exercise) and correctly predictedWoodrow Wilson's election as president. Mailing outmillions of postcards and simply counting the returns, theDigest correctly predicted the victories of Warren Hard-ing in 1920, Calvin Coolidge in 1924, Herbert Hoover in1928, and Franklin Roosevelt in 1932.Then, in 1936, its 2.3 million “voters” constituted ahuge sample; however, they were generally more afflu-ent Americans who tended to have Republican sympa-thies. The Literary Digest was ignorant of this new bias.The week before election day, it reported that Alf Lan-don was far more popular than Roosevelt. At the sametime, George Gallup conducted a far smaller, but morescientifically based survey, in which he polled a demo-graphically representative sample. Gallup correctly pre-dicted Roosevelt’s landslide victory. The Literary Digestsoon went out of business, while polling started to takeoff.Elmo Roper was another American pioneer in politicalforecasting using scientific polls.[1] He predicted the re-election of President Franklin D. Roosevelt three times,in 1936, 1940, and 1944. Louis Harris had been in thefield of public opinion since 1947 when he joined theElmo Roper firm, then later became partner.In September 1938 Jean Stoetzel, after having metGallup, created IFOP, the Institut Français d'OpinionPublique, as the first European survey institute in Parisand started political polls in summer 1939 with the ques-tion "Why die for Danzig?", looking for popular support

or dissent with this question asked by appeasement politi-cian and future collaborationist Marcel Déat.Gallup launched a subsidiary in the United Kingdom that,almost alone, correctly predicted Labour’s victory in the1945 general election, unlike virtually all other commen-tators, who expected a victory for the Conservative Party,led by Winston Churchill.The Allied occupation powers helped to create survey in-stitutes in all of the Western occupation zones of Ger-many in 1947 and 1948 to better steer denazification.By the 1950s, various types of polling had spread to mostdemocracies.

2 Sample and polling methods

Opinion polls for many years were maintained throughtelecommunications or in person-to-person contact.Methods and techniques vary, though they are widely ac-cepted in most areas. Verbal, ballot, and processed typescan be conducted efficiently, contrasted with other typesof surveys, systematics, and complicated matrices beyondprevious orthodox procedures.Opinion polling developed into popular applicationsthrough popular thought, although response rates forsome surveys declined. Also, the following has also ledto differentiating results:[1] Some polling organizations,such as Angus Reid Public Opinion, YouGov and Zogbyuse Internet surveys, where a sample is drawn from a largepanel of volunteers, and the results are weighted to reflectthe demographics of the population of interest. In con-trast, popular web polls draw on whoever wishes to par-ticipate, rather than a scientific sample of the population,and are therefore not generally considered professional.Recently, statistical learningmethods have been proposedin order to exploit Social Media content (such as postson the micro-blogging platform of Twitter) for modellingand predicting voting intention polls.[2][3]

Polls can be used in the public relation field as well. Inthe early 1920s Public Relation experts described theirwork as a two-way street. Their job would be to presentthe misinterpreted interests of large institutions to public.They would also gauge the typically ignored interests ofthe public through polls.

1

Page 2: Opinion Poll

2 3 POTENTIAL FOR INACCURACY

2.1 Benchmark polls

A benchmark poll is generally the first poll taken in a cam-paign. It is often taken before a candidate announces theirbid for office but sometimes it happens immediately fol-lowing that announcement after they have had some op-portunity to raise funds. This is generally a short and sim-ple survey of likely voters.A benchmark poll serves a number of purposes for a cam-paign, whether it is a political campaign or some othertype of campaign. First, it gives the candidate a pictureof where they stand with the electorate before any cam-paigning takes place. If the poll is done prior to announc-ing for office the candidate may use the poll to decidewhether or not they should even run for office. Secondly,it shows themwhere their weaknesses and strengths are intwo main areas. The first is the electorate. A benchmarkpoll shows them what types of voters they are sure to win,those who they are sure to lose, and everyone in-betweenthose two extremes. This lets the campaign know whichvoters are persuadable so they can spend their limited re-sources in the most effective manner. Second, it can givethem an idea of what messages, ideas, or slogans are thestrongest with the electorate.[4]

2.2 Brushfire polls

Brushfire Polls are polls taken during the period betweenthe Benchmark Poll and Tracking Polls. The number ofBrushfire Polls taken by a campaign is determined by howcompetitive the race is and how much money the cam-paign has to spend. These polls usually focus on likelyvoters and the length of the survey varies on the numberof messages being tested.Brushfire polls are used for a number of purposes. First,it lets the candidate know if they have made any progresson the ballot, how much progress has been made, andin what demographics they have been making or losingground. Secondly, it is a way for the campaign to test avariety of messages, both positive and negative, on them-selves and their opponent(s). This lets the campaign knowwhat messages work best with certain demographics andwhat messages should be avoided. Campaigns often usethese polls to test possible attack messages that their op-ponent may use and potential responses to those attacks.The campaign can then spend some time preparing an ef-fective response to any likely attacks. Thirdly, this kindof poll can be used by candidates or political parties toconvince primary challengers to drop out of a race andsupport a stronger candidate.

2.3 Tracking polls

A tracking poll is a poll repeated at intervals generally av-eraged over a trailing window.[5] For example, a weekly

tracking poll uses the data from the past week and dis-cards older data.A caution is that estimating the trend is more difficult anderror-prone than estimating the level – intuitively, if oneestimates the change, the difference between two num-bers X and Y, then one has to contend with the error inboth X and Y – it is not enough to simply take the differ-ence, as the change may be random noise. For details, seet-test. A rough guide is that if the change in measurementfalls outside the margin of error, it is worth attention.

3 Potential for inaccuracy

Polls based on samples of populations are subject tosampling error which reflects the effects of chance anduncertainty in the sampling process. The uncertainty isoften expressed as a margin of error. The margin of er-ror is usually defined as the radius of a confidence intervalfor a particular statistic from a survey. One example isthe percent of people who prefer product A versus prod-uct B. When a single, global margin of error is reportedfor a survey, it refers to the maximum margin of errorfor all reported percentages using the full sample fromthe survey. If the statistic is a percentage, this maximummargin of error can be calculated as the radius of the con-fidence interval for a reported percentage of 50%. Otherssuggest that a poll with a random sample of 1,000 peo-ple has margin of sampling error of 3% for the estimatedpercentage of the whole population.A 3% margin of error means that if the same procedureis used a large number of times, 95% of the time the truepopulation average will be within the 95% confidence in-terval of the sample estimate plus or minus 3%. Themargin of error can be reduced by using a larger sam-ple, however if a pollster wishes to reduce the margin oferror to 1% they would need a sample of around 10,000people.[6] In practice, pollsters need to balance the costof a large sample against the reduction in sampling er-ror and a sample size of around 500–1,000 is a typicalcompromise for political polls. (Note that to get com-plete responses it may be necessary to include thousandsof additional participators.)[7]

Another way to reduce the margin of error is to rely onpoll averages. This makes the assumption that the pro-cedure is similar enough between many different pollsand uses the sample size of each poll to create a pollingaverage.[8] An example of a polling average can be foundhere: 2008 Presidential Election polling average. An-other source of error stems from faulty demographicmodels by pollsters who weigh their samples by partic-ular variables such as party identification in an election.For example, if you assume that the breakdown of the USpopulation by party identification has not changed sincethe previous presidential election, you may underestimatea victory or a defeat of a particular party candidate that

Page 3: Opinion Poll

3.3 Wording of questions 3

saw a surge or decline in its party registration relative tothe previous presidential election cycle.Over time, a number of theories and mechanisms havebeen offered to explain erroneous polling results. Someof these reflect errors on the part of the pollsters; manyof them are statistical in nature. Others blame the re-spondents for not giving candid answers (e.g., the Bradleyeffect, the Shy Tory Factor); these can be more contro-versial.

3.1 Nonresponse bias

Since some people do not answer calls from strangers,or refuse to answer the poll, poll samples may not berepresentative samples from a population due to a non-response bias. Because of this selection bias, the char-acteristics of those who agree to be interviewed may bemarkedly different from those who decline. That is, theactual sample is a biased version of the universe the poll-ster wants to analyze. In these cases, bias introduces newerrors, one way or the other, that are in addition to er-rors caused by sample size. Error due to bias does notbecome smaller with larger sample sizes, because takinga larger sample size simply repeats the same mistake ona larger scale. If the people who refuse to answer, or arenever reached, have the same characteristics as the peoplewho do answer, then the final results should be unbiased.If the people who do not answer have different opinionsthen there is bias in the results. In terms of election polls,studies suggest that bias effects are small, but each pollingfirm has its own techniques for adjusting weights to min-imize selection bias.[9]

3.2 Response bias

Survey results may be affected by response bias, wherethe answers given by respondents do not reflect theirtrue beliefs. This may be deliberately engineered by un-scrupulous pollsters in order to generate a certain resultor please their clients, but more often is a result of the de-tailed wording or ordering of questions (see below). Re-spondents may deliberately try to manipulate the outcomeof a poll by e.g. advocating a more extreme position thanthey actually hold in order to boost their side of the ar-gument or give rapid and ill-considered answers in orderto hasten the end of their questioning. Respondents mayalso feel under social pressure not to give an unpopularanswer. For example, respondents might be unwilling toadmit to unpopular attitudes like racism or sexism, andthus polls might not reflect the true incidence of these at-titudes in the population. In American political parlance,this phenomenon is often referred to as the Bradley effect.If the results of surveys are widely publicized this effectmay be magnified - a phenomenon commonly referred toas the spiral of silence.

3.3 Wording of questions

It is well established that the wording of the questions, theorder in which they are asked and the number and form ofalternative answers offered can influence results of polls.For instance, the public is more likely to indicate supportfor a person who is described by the operator as one of the“leading candidates”. This support itself overrides subtlebias for one candidate, as does lumping some candidatesin an “other” category or vice versa. Thus comparisonsbetween polls often boil down to the wording of the ques-tion. On some issues, question wording can result in quitepronounced differences between surveys.[10][11][12] Thiscan also, however, be a result of legitimately conflictedfeelings or evolving attitudes, rather than a poorly con-structed survey.[13]

A common technique to control for this bias is to rotatethe order in which questions are asked. Many pollstersalso split-sample. This involves having two different ver-sions of a question, with each version presented to halfthe respondents.The most effective controls, used by attitude researchers,are:

• asking enough questions to allow all aspects of anissue to be covered and to control effects due tothe form of the question (such as positive or nega-tive wording), the adequacy of the number being es-tablished quantitatively with psychometric measuressuch as reliability coefficients, and

• analyzing the results with psychometric techniqueswhich synthesize the answers into a few reliablescores and detect ineffective questions.

These controls are not widely used in the polling industry.

3.4 Coverage bias

Another source of error is the use of samples that are notrepresentative of the population as a consequence of themethodology used, as was the experience of the LiteraryDigest in 1936. For example, telephone sampling has abuilt-in error because in many times and places, thosewith telephones have generally been richer than thosewithout.In some places many people have only mobile telephones.Because pollsters cannot use automated dialing machinesto call mobile phones (it is unlawful in the United Statesto use automated dialing machines to reach phones wherethe phone’s owner may be charged simply for taking acall[14]), these individuals are typically excluded frompolling samples. There is concern that, if the subsetof the population without cell phones differs markedlyfrom the rest of the population, these differences canskew the results of the poll. Polling organizations have

Page 4: Opinion Poll

4 5 INFLUENCE

developed many weighting techniques to help overcomethese deficiencies, with varying degrees of success. Stud-ies of mobile phone users by the Pew Research Centerin the US, in 2007, concluded that “cell-only respon-dents are different from landline respondents in impor-tant ways, (but) they were neither numerous enough nordifferent enough on the questions we examined to pro-duce a significant change in overall general populationsurvey estimates when included with the landline sam-ples and weighted according to US Census parameters onbasic demographic characteristics.”[15]

This issue was first identified in 2004,[16] but came toprominence only during the 2008 US presidential elec-tion.[17] In previous elections, the proportion of the gen-eral population using cell phones was small, but as thisproportion has increased, there is concern that pollingonly landlines is no longer representative of the generalpopulation. In 2003, only 2.9% of households were wire-less (cellphones only), compared to 12.8% in 2006.[18]This results in "coverage error". Many polling organi-sations select their sample by dialling random telephonenumbers; however, in 2008, there was a clear tendencyfor polls which included mobile phones in their samplesto show a much larger lead for Obama, than polls that didnot.[19][20]

The potential sources of bias are:[21]

1. Some households use cellphones only and have nolandline. This tends to include minorities andyounger voters; and occurs more frequently inmetropolitan areas. Men are more likely to becellphone-only compared to women.

2. Some people may not be contactable by landlinefromMonday to Friday and may be contactable onlyby cellphone.

3. Some people use their landlines only to access theInternet, and answer calls only to their cellphones.

Some polling companies have attempted to get aroundthat problem by including a “cellphone supplement”.There are a number of problems with including cell-phones in a telephone poll:

1. It is difficult to get co-operation from cellphoneusers, because in many parts of the US, users arecharged for both outgoing and incoming calls. Thatmeans that pollsters have had to offer financial com-pensation to gain co-operation.

2. US federal law prohibits the use of automated di-alling devices to call cellphones (Telephone Con-sumer Protection Act of 1991). Numbers thereforehave to be dialled by hand, which is more time-consuming and expensive for pollsters.

An oft-quoted example of opinion polls succumbing toerrors occurred during the UK General Election of 1992.

Despite the polling organizations using different method-ologies, virtually all the polls taken before the vote, andto a lesser extent, exit polls taken on voting day, showed alead for the opposition Labour party, but the actual votegave a clear victory to the ruling Conservative party.In their deliberations after this embarrassment the poll-sters advanced several ideas to account for their errors,including:

Late swing Voters who changed their minds shortly be-fore voting tended to favour the Conservatives, sothe error was not as great as it first appeared.

Nonresponse bias Conservative voters were less likelyto participate in surveys than in the past and werethus under-represented.

The Shy Tory Factor The Conservatives had suffereda sustained period of unpopularity as a result of eco-nomic difficulties and a series of minor scandals,leading to a spiral of silence in which some Con-servative supporters were reluctant to disclose theirsincere intentions to pollsters.

The relative importance of these factors was, and re-mains, a matter of controversy, but since then the pollingorganizations have adjusted their methodologies and haveachieved more accurate results in subsequent electioncampaigns.

4 Failures

A widely publicized failure of opinion polling to date inthe United States was the prediction that Thomas Deweywould defeat Harry S. Truman in the 1948 US pres-idential election. Major polling organizations, includ-ing Gallup and Roper, indicated a landslide victory forDewey.In the United Kingdom, most polls failed to predict theConservative election victories of 1970 and 1992, andLabour’s victory in 1974. However, their figures at otherelections have been generally accurate.

5 Influence

5.1 Effect on voters

By providing information about voting intentions, opin-ion polls can sometimes influence the behavior of elec-tors, and in his book The Broken Compass, Peter Hitchensasserts that opinion polls are actually a device for in-fluencing public opinion.[22] The various theories abouthow this happens can be split into two groups: band-wagon/underdog effects, and strategic (“tactical”) voting.

Page 5: Opinion Poll

5.2 Effect on politicians 5

A bandwagon effect occurs when the poll prompts vot-ers to back the candidate shown to be winning in thepoll. The idea that voters are susceptible to such effectsis old, stemming at least from 1884; William Safire re-ported that the term was first used in a political cartoonin the magazine Puck in that year.[23] It has also remainedpersistent in spite of a lack of empirical corroborationuntil the late 20th century. George Gallup spent mucheffort in vain trying to discredit this theory in his timeby presenting empirical research. A recent meta-studyof scientific research on this topic indicates that from the1980s onward the Bandwagon effect is found more oftenby researchers.[24]

The opposite of the bandwagon effect is the underdog ef-fect. It is oftenmentioned in themedia. This occurs whenpeople vote, out of sympathy, for the party perceived tobe “losing” the elections. There is less empirical evidencefor the existence of this effect than there is for the exis-tence of the bandwagon effect.[24]

The second category of theories on how polls directlyaffect voting is called strategic or tactical voting. Thistheory is based on the idea that voters view the act ofvoting as a means of selecting a government. Thusthey will sometimes not choose the candidate they pre-fer on ground of ideology or sympathy, but another, less-preferred, candidate from strategic considerations. Anexample can be found in the United Kingdom generalelection, 1997. As he was then a Cabinet Minister,Michael Portillo's constituency of Enfield Southgate wasbelieved to be a safe seat but opinion polls showed theLabour candidate Stephen Twigg steadily gaining sup-port, which may have prompted undecided voters or sup-porters of other parties to support Twigg in order to re-move Portillo. Another example is the boomerang effectwhere the likely supporters of the candidate shown to bewinning feel that chances are slim and that their vote isnot required, thus allowing another candidate to win.In addition, Mark Pickup in Cameron Anderson andLaura Stephenson’s “Voting Behaviour in Canada” out-lines three additional “behavioural” responses that votersmay exhibit when faced with polling data.The first is known as a “cue taking” effect which holdsthat poll data is used as a “proxy” for information aboutthe candidates or parties. Cue taking is “based on thepsychological phenomenon of using heuristics to simplifya complex decision” (243).[25]

The second, first described by Petty and Cacioppo (1996)is known as “cognitive response” theory. This theory as-serts that a voter’s response to a poll may not line withtheir initial conception of the electoral reality. In re-sponse, the voter is likely to generate a “mental list” inwhich they create reasons for a party’s loss or gain in thepolls. This can reinforce or change their opinion of thecandidate and thus affect voting behaviour.Third, the final possibility is a “behavioural response”which is similar to a cognitive response. The only salient

difference is that a voter will go and seek new informationto form their “mental list,” thus becoming more informedof the election. This may then affect voting behaviour.These effects indicate how opinion polls can directly af-fect political choices of the electorate. But directly orindirectly, other effects can be surveyed and analyzed onall political parties. The form of media framing and partyideology shifts must also be taken under consideration.Opinion polling in some instances is a measure of cog-nitive bias, which is variably considered and handled ap-propriately in its various applications.

5.2 Effect on politicians

Starting in the 1980s, tracking polls and related tech-nologies began having a notable impact on U.S. politi-cal leaders.[26] According to Douglas Bailey, a Republi-can who had helped run Gerald Ford's 1976 presidentialcampaign, “It’s no longer necessary for a political candi-date to guess what an audience thinks. He can [find out]with a nightly tracking poll. So it’s no longer likely thatpolitical leaders are going to lead. Instead, they're goingto follow.”[26]

6 Regulation

Some jurisdictions over the world restrict the publicationof the results of opinion polls, especially during the pe-riod around an election, in order to prevent the possiblyerroneous results from affecting voters’ decisions. For in-stance, in Canada, it is prohibited to publish the resultsof opinion surveys that would identify specific politicalparties or candidates in the final three days before a pollcloses.[27]

However, most western democratic nations don't supportthe entire prohibition of the publication of pre-electionopinion polls; most of them have no regulation and someonly prohibit it in the final days or hours until the relevantpoll closes.[28] A survey by Canada’s Royal Commissionon Electoral Reform reported that the prohibition periodof publication of the survey results largely differed in dif-ferent countries. Out of the 20 countries examined, 3prohibit the publication during the entire period of cam-paigns, while others prohibit it for a shorter term suchas the polling period or the final 48 hours before a pollcloses.[27] In India, the Election Commission has prohib-ited it in the 48 hours before the start of polling.

7 See also• Deliberative opinion poll

• Entrance poll

• Everett Carll Ladd

Page 6: Opinion Poll

6 9 REFERENCES

• Exit poll

• Historical polling for U.S. Presidential elections

• List of polling organizations

• Open access poll

• Push poll

• Referendum

• Roper Center for Public Opinion Research

• Sample size determination

• Straw poll

8 Footnotes[1] Cantril, Hadley; Strunk, Mildred (1951). “Public Opin-

ion, 1935-1946”. Princeton University Press. p. vii.

[2] Vasileios Lampos, Daniel Preotiuc-Pietro and TrevorCohn. A user-centric model of voting intention fromSocial Media. Proceedings of the 51st Annual Meetingof the Association for Computational Linguistics. ACL,pp. 993-1003, 2013. http://aclweb.org/anthology/P/P13/P13-1098.pdf

[3] Brendan O'Connor, Ramnath Balasubramanyan, Bryan RRoutledge, and Noah A Smith. From Tweets to Polls:Linking Text Sentiment to Public Opinion Time Series.In Proceedings of the International AAAI Conference onWeblogs and Social Media. AAAI Press, pp. 122–129,2010.

[4] Kenneth F.Warren (1992). “in Defense of Public OpinionPolling.” Westview Press. p. 200-1.

[5] “About the Tracking Polls”. Cnn.com. Retrieved 2013-02-18.

[6] An estimate of the margin of error in percentage termscan be gained by the formula 100 ÷ square root of samplesize

[7] “publicagenda.org”. publicagenda.org. Retrieved 2013-02-18.

[8] Lynch, Scott M. Introduction to Bayesian Statistics and Es-timation for Social Scientists (2007).

[9] Langer, Gary (May 2003). “About Response Rates: SomeUnresolved Questions” (PDF). ABC News. Retrieved2010-05-17.

[10] “Public Agenda Issue Guide: Higher Education - PublicView - Red Flags Public Agenda”. Publicagenda.org. Re-trieved 2013-02-18.

[11] “Public Agenda Issue Guide: Gay Rights - Public View -Red Flags Public Agenda”. Publicagenda.org. Retrieved2013-02-18.

[12] “Public Agenda Issue Guide: Abortion - Public View -Red Flags”. Public Agenda.

[13] “The Seven Stages of Public Opinion Public Agenda”.Publicagenda.org. Retrieved 2013-02-18.

[14] http://transition.fcc.gov/cgb/policy/TCPA-Rules.pdf

[15] Keeter, Scott (2007-06-27). “How Serious Is Polling’sCell-Only Problem?". Pew Research Center Publications.

[16] Blumenthal, Mark (2008-09-19). “More Pollsters Inter-viewing By Cell Phone”. Pollster.com. Retrieved 2008-11-04.

[17] Blumenthal, Mark (2008-07-17). “New Pew data on cellphones”. Pollster. Retrieved 2008-11-04.

[18] Blumberg SJ, Luke JV (2007-05-14). “Wireless Substitu-tion: Early Release of Estimates Based on Data from theNational Health Interview Survey, July–December 2006”(PDF). Centers for Disease Control. Retrieved 2009-06-22.

[19] Silver, Nate (2008-11-02). “The Cellphone effect, con-tinued”. FiveThirtyEight.com. Retrieved 2008-11-04.

[20] Blumenthal, Mark (2008-10-17). “More Cell Phone Datafrom Gallup”. Pollster.com. Retrieved 2008-11-04.

[21] Silver, Nate (2008-07-22). “The Cellphone Problem, Re-visited”. FiveThirtyEight.com. Retrieved 2008-11-04.

[22] Hitchens, Peter (2009). “Chapter 1, Guy Fawkes Gets aBlackberry”. The Broken Compass: How British PoliticsLost its Way. Continuum International Publishing GroupLtd. ISBN 1-84706-405-1.

[23] Safire, William, Safire’s Political Dictionary, page 42.Random House, 1993.

[24] Irwin, Galen A. and Joop J. M. Van Holsteyn. Bandwag-ons, Underdogs, the Titanic and the Red Cross: The Influ-ence of Public Opinion Polls on Voters (2000).

[25] Anderson, Cameron; Pickup, Mark (2010). “10”. VotingBehaviour in Canada. Vancouver: UBC Press. pp. 243–278.

[26] Kaiser, Robert G. (March 9, 2011). “David S. Broder:The best political reporter of his time”. The WashingtonPost. Retrieved 2011-03-09.

[27] Claude Emery (January 1994), Public opinion polling inCanada, Library of Parliament, Canada

[28] Tim Bale (2002). “Restricting the broadcast andpublication of pre-election and exit polls: some se-lected examples”. Representation 39 (1): 15–22.doi:10.1080/00344890208523210.

9 References• Asher, Herbert: Polling and the Public. What EveryCitizen Should Know, fourth edition. Washington,D.C.: CQ Press, 1998.

• Bourdieu, Pierre, “Public Opinion does not exist” inSociology in Question, London, Sage (1995).

Page 7: Opinion Poll

7

• Bradburn, Norman M. and Seymour Sudman. Pollsand Surveys: Understanding What They Tell Us(1988).

• Cantril, Hadley. Gauging Public Opinion (1944).

• Cantril, Hadley and Mildred Strunk, eds. PublicOpinion, 1935-1946 (1951), massive compilation ofmany public opinion polls from US, UK, Canada,Australia, and elsewhere.

• Converse, Jean M. Survey Research in the UnitedStates: Roots and Emergence 1890-1960 (1987), thestandard history.

• Crespi, Irving. Public Opinion, Polls, and Democ-racy (1989).

• Gallup, George. Public Opinion in a Democracy(1939).

• Gallup, Alec M. ed. The Gallup Poll Cumula-tive Index: Public Opinion, 1935-1997 (1999) lists10,000+ questions, but no results.

• Gallup, George Horace, ed. The Gallup Poll; Pub-lic Opinion, 1935-1971 3 vol (1972) summarizes re-sults of each poll.

• Glynn, Carroll J., Susan Herbst, Garrett J. O'Keefe,and Robert Y. Shapiro. Public Opinion (1999) text-book

• Lavrakas, Paul J. et al. eds. Presidential Polls andthe News Media (1995)

• Moore, David W. The Superpollsters: How TheyMeasure and Manipulate Public Opinion in America(1995).

• Niemi, Richard G., John Mueller, Tom W. Smith,eds. Trends in Public Opinion: A Compendium ofSurvey Data (1989).

• Oskamp, Stuart and P.Wesley Schultz; Attitudes andOpinions (2004).

• Robinson, Claude E. Straw Votes (1932).

• Robinson, Matthew Mobocracy: How the Media’sObsession with Polling Twists the News, Alters Elec-tions, and Undermines Democracy (2002).

• Rogers, Lindsay. The Pollsters: Public Opinion, Pol-itics, and Democratic Leadership (1949).

• Traugott, Michael W. The Voter’s Guide to ElectionPolls 3rd ed. (2004).

• James G. Webster, Patricia F. Phalen, LawrenceW. Lichty; Ratings Analysis: The Theory and Prac-tice of Audience Research Lawrence Erlbaum Asso-ciates, 2000.

• Young, Michael L. Dictionary of Polling: The Lan-guage of Contemporary Opinion Research (1992).

Additional Sources

• Walden, Graham R. Survey Research Methodology,1990-1999: An Annotated Bibliography. Bibliogra-phies and Indexes in Law and Political Science Se-ries. Westport, CT: Greenwood Press, GreenwoodPublishing Group, Inc., 2002. xx, 432p.

• Walden, Graham R. Public Opinion Polls and Sur-vey Research: A Selective Annotated Bibliography ofU.S. Guides and Studies from the 1980s. Public Af-fairs and Administrative Series, edited by James S.Bowman, vol. 24. New York, NY: Garland Pub-lishing Inc., 1990. xxix, 360p.

• Walden, Graham R. Polling and Survey ResearchMethods 1935-1979: An Annotated Bibliography.Bibliographies and Indexes in Law and Political Sci-ence Series, vol. 25. Westport, CT: GreenwoodPublishing Group, Inc., 1996. xxx, 581p.

10 External Links• Polls from UCB Libraries GovPubs

• The Pew Research Center nonpartisan “fact tank”providing information on the issues, attitudes andtrends shaping America and the world by conductingpublic opinion polling and social science research

• “Use Opinion Research To Build Strong Communi-cation” by Frank Noto

• National Council on Public Polls association ofpolling organizations in the United States devotedto setting high professional standards for surveys

• Survey Analysis Tool based on A. Berkopec, Hy-perQuick algorithm for discrete hypergeometric dis-tribution, Journal of Discrete Algorithms, Elsevier,2006.

Page 8: Opinion Poll

8 10 EXTERNAL LINKS

Support For Direct Popular Vote

Voter polling questionnaire on display at the Smithsonian Institu-tion

2000 Palm Beach County voting stand and ballot box

Page 9: Opinion Poll

9

Voter Turnout by Race-Ethnicity, 2008 US Presidential Election

Page 10: Opinion Poll

10 11 TEXT AND IMAGE SOURCES, CONTRIBUTORS, AND LICENSES

11 Text and image sources, contributors, and licenses

11.1 Text• Opinion poll Source: http://en.wikipedia.org/wiki/Opinion%20poll?oldid=654413246 Contributors: Vicki Rosenzweig, SimonP, Boud,

Michael Hardy, Isomorphic, 172, Jeandré du Toit, Charles Matthews, Hlavac, Bloodshedder, Chealer, Zandperl, Jxg, Meelar, Gidonb,Timrollpickering, Bkell, Michael Snow, Jord, Srtxg, Jpo, DocWatson42, Duncan Keith, Tom harrison, Xinoph, Jackol, Neilc, Gadfium,Fys, Quota, WOT, Klemen Kocjancic, Cab88, Rich Farmbrough, John FitzGerald, Xezbeth, Bender235, Quinten, Kwamikagami, Sura-chit, Aude, Harryfdoherty, Halavais, Bobo192, Stesmo, Trevj, Grutness, Alansohn, Andrew Gray, Kpwa gok, Avenue, Max Naylor, OlegAlexandrov, Joriki, Woohookitty, Wafry, Junes, Prashanthns, Stefanomione, Lastorset, BD2412, LanguageMan, RadioActive, Thekohser,Fred Bradstadt, Raprat0, Fish and karate, Therearenospoons, Ground Zero, Estrellador*, JdforresterBot, Tequendamia, Bgwhite, ,,n, TheRambling Man, YurikBot, Wavelength, Jlittlet, Lincolnite, RadioFan, Stephenb, Shell Kinney, Rjensen, RazorICE, Nick, Tachyon01,Katpatuka, Zzuuzz, Mais oui!, Whouk, Jeff Silvers, Akrabbim, SmackBot, Ikip, TFMcQ, Stephensuleeman, Delldot, RobotJcb, IsaacDupree, XxAvalanchexX, Chris the speller, Nbarth, Darth Panda, Frsky, Shuki, Yidisheryid, Steven X, AdeMiami, Chrylis, Howard theDuck, Gbinal, DMacks, Kendrick7, BlackTerror, Will Beback, Nelro, Kuru, John, Grumpyyoungman01, Hannah Commodore, RichardF,Levineps, Ft93110, Joseph Solis in Australia, JHP, Chris53516, FakeTango, Ve2jgs, Tawkerbot2, JForget, Americasroof, Daob, Picaroon,Lazulilasher, Korky Day, Mblumber, Gogo Dodo, Epbr123, Dasani, Rethas, Al Lemos, Sandossu, Yettie0711, AntiVandalBot, Mack2,Dylan Lake, Pipacopa, Roving Wordslinger, MER-C, Mcorazao, TAnthony, Schmackity, JamesBWatson, Appraiser, Kawanazii, Recur-ring dreams, Destynova, Alanbrowne, Cgingold, Boffob, Pax:Vobiscum, Votemania, Sk wiki, Maradif, LedgendGamer, Filll, Pollpub,Bob98b3, NRCenter, AntiSpamBot, Flatterworld, Student7, Steel1943, Janice Vian, Sporti, Cmonday, Zigorath, CallieRyan, Una Smith,Lingamer8, PDFbot, David Condrey, Botev, Accounting4Taste, MaesterTonberry, Flyer22, Tartuffoboy, Anchor Link Bot, Melcombe,OneLewis101, Capguru, Jean-Francois Landry, ClueBot, ImperfectlyInformed, Dvorsky, Bcnsubmit, Ottawahitech, Piledhigheranddeeper,Jusdafax, Bruceanthro, G.R.Walden, Qwfp, Kurdo777, Roxy the dog, Jprw, Whooym, Addbot, Baloghmatt, Br1z, Member456345, Poll-ster09, 24Research, WatariSan, Politicas, Patriotis672, Member529, Optimistic5, Optimel4, Member564, Lincolninst, Streetpav, Stafaxis,English Linesman, Emostafa 2008, Bhentze, Yobot, Jonathan26, AnomieBOT, Materialscientist, JohnFromPinckney, Quebec99, LilHelpa,Cydelin, Wikipersonae, Sophus Bie, CaptainFugu, Tobby72, Math321, Citation bot 1, Blargh29, Pinethicket, Kiefer.Wolfowitz, Drcrnc,Rplal120, Neil Chadda, Cnwilliams, Ukr-Trident, Lotje, January, Jiffles1, RjwilmsiBot, MMS2013, NameIsRon, Polly Ticker, York-shireKat, John of Reading, Ajraddatz, Maccabi72, Vladimir.qq, Sungzungkim, H3llBot, Quae legit, Erianna, Natanv, Donner60, Ip82166,ClueBot NG, Widr, Dumebi1986, MerlIwBot, Trift, Helpful Pixie Bot, Machaven, Bbzzme, Okstacie123, Redroar75, Conifer, Aisteco,BattyBot, Tutelary, Pratyya Ghosh, Etp01, Mjh110101, Mrb08876, SD5bot, Rebskin, Futurist110, Maniesansdelire, Mmqa, JaconaFrere,Skr15081997, Monkbot, Chaz.bush1, Avinashmjpatil, Gutinapo, MoreInformativePolls and Anonymous: 232

11.2 Images• File:Ambox_globe_content.svg Source: http://upload.wikimedia.org/wikipedia/commons/b/bd/Ambox_globe_content.svg License:

Public domain Contributors: Own work, using File:Information icon3.svg and File:Earth clip art.svg Original artist: penubag• File:Commons-logo.svg Source: http://upload.wikimedia.org/wikipedia/en/4/4a/Commons-logo.svg License: ? Contributors: ? Originalartist: ?

• File:Folder_Hexagonal_Icon.svg Source: http://upload.wikimedia.org/wikipedia/en/4/48/Folder_Hexagonal_Icon.svg License: Cc-by-sa-3.0 Contributors: ? Original artist: ?

• File:Mergefrom.svg Source: http://upload.wikimedia.org/wikipedia/commons/0/0f/Mergefrom.svg License: Public domain Contributors:? Original artist: ?

• File:People_icon.svg Source: http://upload.wikimedia.org/wikipedia/commons/3/37/People_icon.svg License: CC0 Contributors: Open-Clipart Original artist: OpenClipart

• File:Portal-puzzle.svg Source: http://upload.wikimedia.org/wikipedia/en/f/fd/Portal-puzzle.svg License: Public domain Contributors: ?Original artist: ?

• File:Support_For_Direct_Popular_Vote.png Source: http://upload.wikimedia.org/wikipedia/commons/f/fa/Support_For_Direct_Popular_Vote.png License: Public domain Contributors: Own work Original artist: Szu

• File:TallahaseePalmBeachBallotBox1.jpg Source: http://upload.wikimedia.org/wikipedia/commons/9/94/TallahaseePalmBeachBallotBox1.jpg License: CC BY 2.5 Contributors: Own work Original artist: Infrogmation

• File:Voter_Turnout_by_Race-Ethnicity,_2008_US_Presidential_Election.png Source: http://upload.wikimedia.org/wikipedia/commons/f/fd/Voter_Turnout_by_Race-Ethnicity%2C_2008_US_Presidential_Election.png License: CC BY 3.0 Contributors: Ownwork Original artist: Rcragun

• File:Voter_poll.jpg Source: http://upload.wikimedia.org/wikipedia/commons/b/bf/Voter_poll.jpg License: CC BY-SA 3.0 Contributors:Own work (Original text: I (RadioFan (talk)) created this work entirely by myself.) Original artist: RadioFan (talk)

• File:WMF_Strategic_Plan_Survey.svg Source: http://upload.wikimedia.org/wikipedia/commons/6/69/WMF_Strategic_Plan_Survey.svg License: CC BY-SA 4.0 Contributors: <a href='//commons.wikimedia.org/wiki/File:WMFstratplanSurvey1.png' class='image'><imgalt='WMFstratplanSurvey1.png' src='//upload.wikimedia.org/wikipedia/commons/thumb/e/e7/WMFstratplanSurvey1.png/100px-WMFstratplanSurvey1.png' width='100' height='83' srcset='//upload.wikimedia.org/wikipedia/commons/thumb/e/e7/WMFstratplanSurvey1.png/150px-WMFstratplanSurvey1.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/e/e7/WMFstratplanSurvey1.png/200px-WMFstratplanSurvey1.png 2x' data-file-width='1100' data-file-height='910' /></a> Originalartist: Fred the Oyster

• File:Wiki_letter_w_cropped.svg Source: http://upload.wikimedia.org/wikipedia/commons/1/1c/Wiki_letter_w_cropped.svg License:CC-BY-SA-3.0 Contributors:

• Wiki_letter_w.svg Original artist: Wiki_letter_w.svg: Jarkko Piiroinen

Page 11: Opinion Poll

11.3 Content license 11

11.3 Content license• Creative Commons Attribution-Share Alike 3.0