Political Polling Common Errors

Political polling in the United States can be a dicey subject. While there are proven methods for getting fair and unbiased results in polling, in the United States citizens are almost never given these results freely. As most of the ‘infotainment’ networks out there fund, produce, and advertise their own polls that serve the agendas of the parties they support, we as Americans are forced to do our own research and come up with our own methods for finding unbiased polling. While it’s not a terrible prospect for Americans to do their own research on candidates and issues (in fact it is ideal for a true working democracy), I think we all know that the sad truth is that most Americans will not be doing this research any time soon, and instead will be relying on the information provided to them on cable news channels. But I digress… In this blog I will be discussing common polling errors and biases, including those involving sampling, survey construction, and even campaign “push polling,” and hopefully showing some good examples of all of these errors in practice.

An easy glance at the president’s job approval rating polls on pollster.com (http://www.pollster.com/polls/us/jobapproval-obama.php) reveals that two of the leading polling organizations, Gallup and Rasmussen, have overwhelmingly incompatible numbers when it comes to President Obama’s approval rating, with Gallup showing a 52% approval rating against only a 38% disapproval rating, and Rasmussen showing a 47% approval rating against a staggering 52% disapproval rating. Let’s say for a second that it would be unfair to assume that either one of these polling organizations is affiliated with or supports one political party or the other. It’s a fair assumption, as neither one has ever had visible problems with push polling or reports of campaign financing. This being the case, the problem with the discrepancy between these numbers could have been in the sampling of the polls. Researchers try to get a sample size that is substantial enough to get a sampling error of plus or minus 4% or less (Hollihan, 174). Since the difference between these numbers is much higher than that, perhaps one or the other of these organizations did not get a sufficient sample size. However, this seems unlikely since both organizations are professional pollsters and as such are well-versed in the mathematics of sample sizes. A more likely explanation is that the sample sizes that were obtained by one, the other, or both groups were not accurate representatives of the American public. It is possible that despite precautions and efforts to give everyone in the population an equal chance of being polled, that the sample was skewed by large numbers of certain groups (ibid, 174). Perhaps for example, Rasmussen received a disproportionately large number of rural respondents, leading to a more unfavorable viewing of the president. Conversely, perhaps Gallup got a lot of respondents from urbanized neighborhoods or a lot of African-American respondents, making their viewing more favorable of the president.

According to Hollihan, common problems with survey questions include oversimplification of issues; loaded, biased and unclear questions; and making inferences of the results of a poll for questions that were never asked (ibid, 175). Recently in Virginia, a poll was conducted to see if Virginians were in favor of closing the ‘Gun Show Loophole,’ with the pro-gun state falling surprisingly on the side of closing the so-called loophole (http://twitter.com/ProtestEasyGuns/status/4968881627). Unfortunately, the question in the poll said very little about what this loophole actually was. The problem was that people see the word ‘loophole’ and automatically assume that there is something illegal or shady going on. The ‘loophole’ in this case involves a buyer buying from an ‘unlicensed dealer.’ This is perfectly legal under state law, and selling guns at a gun show is one of the safest ways to sell firearms, but the word ‘unlicensed’ in this case once again leads those being polled that something illegal is going on, or that the dealer is in some way shady or untrustworthy. The survey question in this way has a large amount of loaded language, and one could argue also that despite the appearance of a clear question, what the inquiry leaves out makes the question’s language more opaque than what it ought to be. Hollihan uses two similar examples of the Nixon and Clinton impeachments, in both cases showing how those being polled often did not understand what they were being asked. In the case of Nixon, the word ‘impeachment’ did not agree with people, but when asked about the trial of Nixon without the operative word in the question, they were in favor of it (Hollihan, 175-176). In Clinton’s case, some members of Congress voted for impeachment, but didn’t necessarily want him removed from office, further muddying the issue for voters (ibid, 176).

Political “push polling” is another way of getting inaccurate results in polling. With sampling problems and even survey question problems, often times the mistakes made that create inaccurate polls are just that: mistakes. The same is not true for “push polling.” Push polling is the deliberate attempt by a pollster (or campaign posing as a pollster) to influence someone under the guise that they are simply conducting an honest poll (Hollihan, 177). A recent example of this is a poll in Ohio that was conducted over the phone during the 2008 presidential election. The pollster would ask the persona series of ‘agree/disagree’ questions about Barack Obama. Sample questions are as follows: “agree/disagree that Obama is too inexperienced to be President,” “agree/disagree that Obama is too risky of a choice for President,” “agree/disagree that Obama as President would pose a threat to our national security;” and you get the idea from there (http://www.huffingtonpost.com/2008/09/11/nasty-anti-obama-push-pol_n_125607.html). You see, even though the questions gave the person the chance to agree or disagree, every question is framed in a negative way towards the candidate. This particular poll did not care about peoples’ responses at all; rather it was intended to make people who planned on voting for Obama re-think their vote and perhaps go with a different candidate instead. Hollihan recounts a similar example of push polling by the Bush campaign in 2000, one that asked agree/disagree questions framing John McCain’s stances on key issues in a negative light. This campaign took it even a step further though, asking McCain supporters (in the South no less) if they would re-think their vote if they knew McCain had fathered an illegitimate black child (McCain has an adopted Bangladeshi daughter) (Hollihan, 178-179). This inference put the Senator’s moral credibility into question in a key demographic, and who knows if that played a major role in George W. Bush’s victory in the Republican Primaries that year.

As you can see, nothing in politics can ever be easy and cut-and-dry, and political polling gets very complicated. Biased pollsters are always looking to sway votes; reckless amateurs can often throw off accuracy through hastily written and poorly thought out questions; and even when we get an accurate and unbiased polling organization, sampling can cause even the most neutral of polls to become skewed if the conditions are at all wrong, resulting in vastly different numbers for the same polls done by different organizations. What we must continue to try and do is work to reduce the number of errors and biases in polls in an effort to keep polls relevant in today’s political landscape. In order to do so, as Americans we must reject the polls fed to us by television and radio infotainment, and instead do our own research and find our own sources polling that we find to be reliable and accurate for all aspects of the political spectrum.

Works Cited

Hollihan, Thomas A. Uncivil Wars: Political Campaigns in a Media Age. 2nd ed. Boston: Bedford/St. Martin's, 2009. Print.

"PE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd" //1){setTimeout(function(){onCondition(D,C,A-1,B)},B)}}};//]] / Protest Easy Guns: 8 out of 10 Virginians wan ..." Twitter. Web. 27 Oct. 2009. .

"Pollster.com: National Job Approval: Pres. Barack Obama." Pollster.com - Political Surveys and Election Polls, Trends, Charts and Analysis. Web. 27 Oct. 2009. .

Stein, Sam. "Nasty Anti-Obama Push Poll Launched In Ohio." Breaking News and Opinion on The Huffington Post. 11 Sept. 2008. Web. 27 Oct. 2009. .

Infotainment and Political Advertising

For this week’s blog, the prompts I was forced to choose between were “Political News” and “Political Advertising.” I smirked a bit when I saw these, as I consider the two topics to be chief contributors to the problem of voter ignorance and misinformation on the true issues. So in trying to find a good source to write this blog, I figured I would look to what I consider to be the two biggest names in punditry; FOX News (http://www.foxnews.com/) and MSNBC (http://www.msnbc.msn.com/). Leading the charge in today’s political “news,” or what I like to call ‘info-tainment’ (used from here on out without quotations and without the hyphen in the middle, as its use here is mainly to emphasize the two words it combines), the websites for these two entities offer a look at the major political news sources, which conveniently enough double as political advertising for the respective political parties that they insist they do not favor. As you will see, a brief trip around even just the home pages for these two sites and the stories attached to them can tell us a great deal about the key platforms and issues of the Democratic (MSNBC) and Republican (FOX News) Parties. As almost everybody knows, these outlets are both also television networks, which along with their websites allows them to reach an astounding number of people, especially now since the average American spends more waking hours watching TV than they do communicating interpersonally (Hollihan, 108).

The first thing that caught my eye on the two sites was the issue of health care. No issue has been more widely discussed over the past several weeks than health care, and specifically the bills that have been written and re-written over the past few months in an effort to reform the system. Surprisingly enough, FOX only has one story about health care reform on the front page of their website, compared to five links to stories about health care on the MSNBC. Furthermore, the language that is used in headlining the articles on health care is worth noting. On FOX, the headline can be read as “Dems Pile on Health Insurers over Critical Report.” This headline sends very specific communications messages. First, using the phrase ‘critical report’ is an attempt to remove responsibility from the health insurers, inferring that the criticalness of the report was outside of their control, when in reality there were probably some things the insurers could have done differently if they wanted to receive a better review. Some may consider that to be a bit of a stretch, but the other part of the headline I feel is even more obvious language. The term ‘pile on’ almost universally in the English Language carries a negative connotation with it. For a conservative news source to say the liberals are ‘piling on’ a group that they are also painting as a victim in the headline almost certainly is trying to send a message. Similarly, the liberal MSNBC uses strategic wording in their article headlines designed to make Republicans look bad. One headline suggests that a Republican senator has ties to a major insurer (insurers are evil on this site). When one reads the article, the facts supporting this claim are severely lacking, but the headline is what most people see first and how many will form their opinion, especially those who choose not to read the whole article. Another article praises a man for treating any and all patients that come to him without any form of health insurance, but does not once mentions the repercussions of doing so or the burden that it could potentially cause tax payers.

But health reform is not the only issue where these two sites show their support along party lines. On the FOX News site there are two separate articles discussing gun rights and one touching on immigration reform. On MSNBC’s site, one can find articles about gay rights and global climate change. What these articles are really doing is outline the top issues for the parties that each of these infotainment sites ‘unofficially’ represent. Democrats like to push the agenda of global warming and gay rights while the Republicans, ever conscious of homeland security like to keep the second amendment front and center along with the issue of border patrol. This is where in America we see news networks doubling as political advertising. They build up their own sides’ positions on their favorite issues while doing their best to malign the other party’s, or dismiss them altogether, as is the case with many of these articles (they don’t even mention the other sides’ position in most of them).

But perhaps the most issue that was the most telling (and the most interesting) of the issues that revealed these two infotainment giants’ political alignment is their stances towards the ongoing military action in Iraq and Afghanistan. All-told, there were no less than five stories on each of the two sites’ front pages about the war on terror and military action across the Middle East. What I found interesting about this was the way that the two sites aligned themselves. FOX’s stories featured almost universal negativity when discussing the events going on overseas. Bleak headlines focused mainly on terror-related deaths and stories that seemed to be suspicious of the Obama administration’s plans to continue the same military actions that they condemned Republicans for in order to get elected last fall. Oppositely, MSNBC’s view painted a picture of an administration making careful decisions in the Middle East and having open minds to new ideas on how to handle the current military climate overseas. What I find interesting and ironic about the take on both sides is the obvious flip-flop from the staunch positions they had during the previous administration, where the liberal side of infotainment endorsed the notion of getting completely out of the Middle East as quickly as possible while the conservative side advocated staying as long as it took.

This kind of hypocrisy promotes the kind of partisanship that both Republicans and Democrats in office claim to be trying to transcend, yet neither party seems to be making much of an effort to distance themselves from these infotainment outlets. I’m sure their reasoning behind it is to allow these networks to attack their opponents for them, so as not to expose themselves to criticism for running negative ‘attack ads’ (Hollihan, 158). Furthermore these news media sources are doing more than ever to keep themselves connected to their respective parties. MSNBC uses phrases in their headlines and articles like ‘the administration’ and ‘the GOP’ while words like ‘Democrats’ or ‘liberals’ are rare to come up. The reason is that they are trying to act like they are on an informal basis with the Democratic Party and their language is geared to suggest that they are simply ‘one of them.’ FOX does the exact same thing, calling out the ‘Obama Administration’ and ‘Democrats’ while almost never mentioning ‘the GOP,’ ‘conservatives,’ or ‘Republicans.’

After all of this, it is easy to see how and why political news and political advertising are so intertwined. These news outlets are two of the biggest that the American public turn to for their information about politics, and yet even a simple glimpse around their websites reveal that they are little more than political advertisements for the key issues of the parties they support, in effect ‘priming’ voters into seeing in the news again and again what they think that the public should be voting on come election time (Hollihan, 120). The beauty of this form of political advertisement is that it can have the effect of bypassing Kathleen Hall Jamieson’s “news-room guide” for thinking critically about political ads, mostly because most people reading into these infotainment outlets don’t realize that they are advertisements (Hollihan, 165). One can only hope that the continued growth and access to technology will help Americans in the future to be able to differentiate between news and infotainment political advertising, and that in the future there will be more credible news sources that have as easy access as these two media giants.

Class Readings Without Provided Link Above

Hollihan, Thomas A. Uncivil Wars: Political Campaigns in a Media Age. 2nd ed. Bedford/St. Martin's, 2008. Print.

Dishonesty and Candidate Image

Those who know me well can attest to my disdain for political rhetoric, which I would argue is more of a form of sophistry in the absence of real knowledge of a subject than it is the useful tool for ‘informing’ the public that most politicians would claim it is. The easiest place to see this political chest-pounding is in political elections, where smooth words are supposed to make us ignore lack of experience, and war stories are supposed to help us ignore a lack of knowledge or innovation to inject into a national problem (either of these seem familiar?). But for lack of reliable and objective sources focusing on recent campaigns and elections, I have instead decided to focus this blog on the vague and deceptive language of candidate speechwriting and the harm that this kind of communication brings to the American people. I will explore several speeches, point out some very obvious problems with each speech’s wording, and speculate on the ramifications of this communication.

The first and most recent example that I could find of vagueness and deception in political speeches was President Obama’s address to the nation on health care. I will reference paragraphs of the speech (as they are divided in the text provided), the full text of which can be found here: http://www.huffingtonpost.com/2009/09/09/obama-health-care-speech_n_281265.html. Most specifically, in paragraphs 26 and 27 the president outlines his plan to require all individuals to have health insurance, pointing out that for anyone not to under his plan would be simply ‘irresponsible.’ What he fails to mention is that any person without insurance under this plan will be fined between $750 and $3800 depending on income levels (Individual Responsibility section of the framework), simply for not having insurance or buying into his public option (text of the framework of the bill available here: http://thehill.com/images/stories/news/2009/september/090809/baucus_health.pdf). This may not seem like a very big deal until you look at it from a situational aspect. Aside from the obvious dishonesty of this in the fact that Obama campaigned heavily on the stand that he would not make insurance mandatory, say for a second that a man loses his job, a job that gave him health benefits. If that man does not find a new job very quickly, he is either forced to pay an enormous fine for not having insurance, or buy into the public option, whether he wants to or not. Thus, this health care plan is designed to eventually kick more and more people into the public option, eventually giving it the majority that it needs to do away with private insurance altogether. Obama is careful to conceal this, as he knows that the American people would be opposed to the final goal of this plan, but he knows that very few people will catch his wording, and those that do and complain about it can easily be dismissed as ‘crazy pundits’ or ‘racists,’ if you ask Jimmy Carter. The result, in this case, could end up being a vote for a policy that if you didn’t take the time to read yourself (most people don’t have the time, heck most of Congress hasn’t read the whole thing), maybe you don’t believe in after all.

I don’t mean to just pick on President Obama either. If Republicans think that their political figures don’t participate in this type of political deception, they need to pay look more closely. I turn my attention now to a speech given by the previous president, George W. Bush, given September 24, 2008 addressing the nation’s economic crisis (Full Text: http://www.nytimes.com/2008/09/25/business/worldbusiness/25iht-24textbush.16463831.html). After his initial remarks, Bush is very quick to point out that these financial problems have been developing over a long period of time (paragraph 12). By paragraphs 15 through 18 he is now coming to what he feels is the crux of the problem in the housing market, lenders giving home loans to people who could not afford to pay them. The dishonesty here is very simple; the president of the United States is showing no responsibility for the problems of the country. He points out that these problems started a long time ago to make sure that people know that he was not responsible for how this crisis started. His comments in paragraphs 15-18 hint at where he places the blame, a bill passed by president Clinton during his administration, a bill that Bush feels led to lenders lending to people who they knew could not afford to pay them back (article outlining the bill found here: http://www.nytimes.com/1999/09/30/business/fannie-mae-eases-credit-to-aid-mortgage-lending.html?n=Top%2FReference%2FTimes%20Topics%2FSubjects%2FM%2FMinorities%20(US)). Whether or not these claims by Bush are true are irrelevant, the simple fact is that the man who had resided over the U.S. for most of eight years took no responsibility for the financial crisis that had the nation on the verge of economic depression. He instead blamed someone else, without naming them (so as not to sound to the American public that he was blaming someone else) and hid his lack of knowledge behind free-market rhetoric. On the second page of his speech (paragraph 6 on page 2), Bush states that his inactivity during this time of crisis can be attributed to his belief in “free enterprise.” Sounds to me like the reality of the situation was that Bush had no idea what to do to help the economy at that time. In this case, deception was more about candidate image than anything else, but the harm caused by this type of deception can lead to elections where the image of a candidate can be perceived unjustly as the image of the entire party, which is where public voice, “the ways citizens give expression to their views on public issues” (Zukin, 54).

Deception and careful wording can be found during the reign of the president before Mr. Bush as well, this time in a slightly different context. During President Clinton’s impeachment proceedings, his careful wording kept him from losing a perjury trial, and not even the lawyer who was interviewing him as a witness caught on to the deception. The first and most famous example of this was his notorious “I did not have sex with that woman!” line. The deception in this is obvious to anyone who has paid attention to the news; he never had intercourse with Monica Lewinsky, but they did engage in oral sex. Perhaps the lesser known of Clinton’s carefully-worded statements was his claim that he was never alone in the Oval Office Complex with Lewinsky, which was true. The problem was that the question that was asked of him was whether he had ever been alone in the Oval Office with her. The fact that he was never alone in the Oval Office Complex with her only means that the Secret Security Guards charged with guarding Clinton were doing their job and standing in the Oval Office Complex just outside the doors into the Oval Office. Once again, candidate image was the main reason for this deception, and the trust of the American people was influenced greatly by the truth that came out later in the Clinton scandal, and the trust that was lost may very well have cost Al Gore and the Democrats the election of 2000. This is an example of the phenomena of the ‘Post-Watergate Era,’ where media scrutiny on candidates’ character forces candidates to go along with certain social norms and values that appease the general public (Hollihan, Ch. 11).

The purpose I have in writing this blog goes back to the lyrics of the old The Who song “Won’t Get Fooled Again,” ‘meet the new boss, same as the old boss.’ There are countless more examples of political dishonesty in the U.S. from aldermen to senators to the president himself. Almost every politician in the U.S. today is inherently dishonest, and if you think some new guy coming in is any different, you are just fooling yourself. Honesty is identified by many researchers as an important characteristic that voters take into consideration when deciding which candidate to vote for (Hollihan, Ch. 4). The problem is that not enough people are identifying this dishonesty. By studying the words of candidates’ speeches and responses to questioning, you can often suspicious rhetoric, and it is then more than ever that you must do your research, study a candidate’s record, and decide for yourself whether, if elected, this person’s promises will hold up, or if past actions speak louder.


Readings from Class without Provided Links


Hollihan, Thomas A. Uncivil Wars: Political Campaigns in a Media Age. 2nd ed. Bedford/St. Martin's, 2008. Print.

Zukin, Cliff. A New Engagement: Political Participation, Civic Life, and the Changing American Citizen. New York: Oxford UP, 2006. Print.