Polls and Power

Statistics = Power.

As we have already discussed on the Fodder, we live in a modern America that thrives on hot button issues, attacks and diatribes. All of these are inevitably propped up by statistics cut from the cloth of conservatism, liberalism and every other conceivable “ism” under the Sun. So where do politicians, lobbyists and activists get these statistics? One word: polls. In life, statistics are power. Secularity demands it. Although it is an overwhelmingly positive progression in society, there still lie serious pitfalls in this mindset. The most prominent is the unfortunate reliance on pollsters. The vast majority of pollsters are performing a vital function of civil society. Not all are, however. What happens when a polling organization goes rogue?

Luckily, we happen to have a lovely little case study unfolding before out eyes. The Daily Kos, a liberal political blog, recently filed a law suit alleging that the polling firm, Research 2000 (R2K), skewed and possibly fabricated statistics used in Daily Kos stories. For over a year, the Daily Kos ran weekly opinion polls administered and researched by R2K. After R2K received a terrible ranking in comparison to other polling firms in early June, the Daily Kos terminated its partnership with the firm. Soon afterward a group of well-known, reliable pollsters came forward to question the overall validity of the numbers R2K was producing. Submitting a preliminary analysis of R2K’s methodological results, the group found there there was almost no statistical chance that the numbers produced by R2K were reliable. In turn, Markos Moulitsas (founder of the Daily Kos) filed the lawsuit referenced above. R2K vehemently denies any wrong doing. (For more information, Pollster.com ran an informative and scathing piece on the case this past weekend.) While this is clearly an extreme example of probable statistical tampering, it still shows the relative ease in which a person or group can change the way others argue and back up their claims. It is a growing and unnerving trend in today’s political landscape.

This is not the first, nor the last, discussion that will be had about the accuracy of polling in America. Case and point: the 2008 Presidential election. From the McCain campaign staking claims on questionable polling data to sweeping analysis of the pollster landscape, the 2008 election was riddled with whispers of inaccuracies and discrepancies. While these whispers never turned into anything meaningful, it is important remember that they were there. Furthermore, this discussion is not restricted to the 2008 Election. Who can forget the media travesty of 2004? Why did the major news networks erroneously call the election in favor of Senator John Kerry? Exit polling. Admitting flaws in 2005, these exit polls led a myriad of people, including former UK Prime Minister Tony Blair, to believe Senator Kerry won the Presidency. We all know the story of what ensued after the election night controversy.

The point of bringing this developing story to you is this: be careful where you get your statistics and of seemingly problematic numbers. Before you use them to make an argument, solidify a research paper or impress a member of the opposite sex, make sure they are from a reliable source. Next time you see a graph in USA Today, on Foxnews.com, in Time, on CNN or in a textbook; check the fine print. Google the firm that is providing the numbers for that chart. It might give you pause. Always see where the numbers are coming from because in this world statistics are power.

  1. No comments yet.
  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: