Let's take a poll...


We hear it often in class, "Science cannot prove anything," and during last class as I was listening to our guest speaker ask us if we trust the experts and what they say, or if we look into things ourselves I couldn't help but question why we so easily accept what we are told.



picture source: http://www.andregarde.com/wp-content/uploads/2012/04/adaptivebacklinking-60pctofthetime.png

With that being said and Election Day fast approaching, I'd like to draw your attention to polling. Politics is sadly not the most interesting topic for a lot of people. It seems to be proven time and time again that people adopt the views of their parents, or use quick cues like their party identification to decide whom they support instead of looking into things on their own. This is true in many aspects of our lives: the acceptance of things based on accessibility rather than knowledge.


We have all heard the debates and statistics being thrown at us as we try and flip through the channels on our television. This video (http://www.colbertnation.com/the-colbert-report-videos/267153/march-11-2010/the-colbert-repoll---scott-rasmussen) from the Colbert Report is comedic, but asks a truly interesting question, what does polling actually do? What does it prove? Is it just another heuristic used to draw our attention and create an easy association for what we think we believe?


There is a comprehensive study (http://www.freerepublic.com/focus/f-news/1004883/posts) that provides a wide range of answers to various questions surrounding whether or not polls are accurate, but in essence, it states that they are not. The study was conducted in California and published on this website in 2003. It analyzed 20 polls and 100 individual results. The first question and most broad it sought to answer "Are political polls accurate?" The answer seemed to be that most polls are not. Even factoring in the Margin of Error, polls really don't tell us the facts.


While it may be true that polls do not tell us the facts, the introduction to this study raises a few more questions for me. At that start of the article, the author talks about the media telling us what we think, about the ignorant acceptance of this hardly accurate data, and how sadly, politicians act on the results. In my opinion, there are several third variables that come along with opinion polls, and these variables may play a huge part in why polls seem to be inaccurate.


Opinions are always changing. They are linked to moods and attitudes as well as many other things. In my social psychology class, we learned about persuasion and attitude change. According to the Yale Attitude Approach it all depends on the "Who, What, and To Whom." This means, who is saying it (telling you what to think), what they are saying, and the audience. There are also two ways to change someone's attitude, either centrally or peripherally. If you are centrally doing so, the audience must be motivated and willing to listen, they have to actual take in the "what" and focus on the argument rather than the "who," or the person saying it. Peripherally changing an opinion is much more superficial, it is taking the easy way out as I have previously stated so many of us do. We care more about the "who," factoring in their attractiveness or whether or not they are deemed an "expert," as opposed to what they are saying. Again, the audience's motivation matters, and for politics is a sad truth that many people are unmotivated to find their own truth, perhaps that is why polls are relied on, but also why they are constantly changing. Peripheral attitude change only lasts for the short term.


Another article (http://knowledge.wharton.upenn.edu/article.cfm?articleid=1843) questioning the accuracy of polling from the University of Pennsylvania factors in the way polling is implemented i.e. Cell Phones versus Landlines. It is important to consider the kind of person who has a landline as their main source of communication versus a cell phone: their socioeconomic status, what area of the country they live in and the ideology associated with these things.


So again I am drawn to wonder what we can ever consider to be certain. In Science, we have learned nothing can be proven, and perhaps this carries on further into our daily lives. We accept the expert opinions because they come from the experts, we consider the attractiveness of the speaker, but do we actually listen to what is being said? Will you change your behavior now that this issue has become salient to you? Either way, I hope you now question the accuracy of polling and all things you are told you should believe.


Psychology Text Book Source:

Aronson, Wilson, & Akert, Social Psychology, 8th edition.


It is very interesting that so much trust is often put into polls when they are often so inaccurate. When polled, many people often put no thought into what they answer. I know I am even guilty of it. After a long day of school or work, the last thing you want to do is spend time filling out a poll. Many people give bogus and thoughtless answers and others simply choose not to be polled at all. In a review of polling practices, the Roper Center of the University concluded that there were three types of distinguishable polling errors, sampling, measurable, and no-response that can be discounted through an error equation. Although this will sort out some polling problems, it does not solve the problem of thoughtless polling answers. How can a poll be reflective of the population when people really don't think that way?
Here's a link to the Roper Center site: http://www.ropercenter.uconn.edu/education/polling_fundamentals_error.html#.UJdQ1WnwKFc

Polls are accurate as far as they represent the answers provided by their respondents.People lie to others and to themselves. People don't know and will not admit it, I think people feel pressure answering questions, regardless of how deliberate bias. People are vain, cynical, flippant, etc.It's been demonstrated many times that people lie, consciously or not, about the food they eat, about their driving habits, and so on.
Furthermore, I think the inaccuracy of polls has to do with the conscious or unconscious lies of the respondents. Yes, a poll can't ever be 100% accurate unless every single subject(in this case voters) are asked. But maybe even then people will change their minds or decide not to vote, causing the expected winner the election. Therefore I think in most polls, the systematic errors are bigger than the statistical errors.

Mayra, you raise an interesting point that again relates to the psychology of opinions and the actions of people. Opinion polls are inaccurate because of their instability in nature. The theory of cognitive dissonance somewhat explains why people lie or do not act in the ways that they believe is "right" and moral. The theory is explained
here, it states in essence, that when someone's action does not match up with their cognitive beliefs, they do something to reduce the discomfort and dissonance. In voting, I think this could apply to: you know it is right both literally and figuratively to vote, but if you don't, you may make excuses or claim neither candidate is appealing to reduce your cognitive dissonance and discomfort. This can perhaps effect polling and the may third variables that change opinions and actions in an unstable and constantly changing political environment.

Leave a comment

Subscribe to receive notifications of follow up comments via email.
We are processing your request. If you don't see any confirmation within 30 seconds, please reload your page.

Search This Blog

Full Text  Tag

Recent Entries

Everyone has heard of them as being the best car out there, mainly cause of gas prices. Hybrids are sweeping…
People everywhere are breaking up, just in time for the holidays. And the more couples I see parting ways, the…
Pregnancy Tests
While browsing Andrew's blog and looking to see all of the posts that I missed (I'm pretty sure I haven't…

Old Contributions