I heard yet another statistic today. It’s ridiculously off base from last week’s skewed stat. Last week, 90% of ALL Americans favored gun control. This week… 64%!!!! Wow! Statistically speaking (heehee) that is way off from last week’s prognosticators spewing numbers with furrowed brows that 9 out of 10 Americans want background checks. But are Pro-Gunners rejoicing at the sudden decline in Anti-Gunners? No, they are not. The reason? Everyone knows that polls and statistics sited by politicians and the Media are lies used to legitimize their position.
Back in the day, statisticians were relied upon to show unbiased looks at a given problem. How did they do that? Numbers. Numbers don’t care… They are what they are. How did one attain the numbers to get an unbiased view of something? Polls.
I’ve been hearing polls sited and stats brandished heavily since just before the previous election that I had to figure out what they were all about. What I found disgusted me. It seems that anything touched by politicians and journalists becomes as foul and tainted as yesterday’s bran muffin passing through your system.
Marketing curriculums in universities around the world include a cautionary tale about polls in the lesson of the 1936 Literary Digest Poll. They use this as an example of what NOT to do. In 1936, the periodical The Literary Digest polled all of its readership with who they favored to be the next president. A staggering 2.4 million respondents (compared to an average of 1200 for a typical poll today) tipped Republican Alf Landon to win the presidential election over Democrat Franklin D. Roosevelt. Statistically, this was essentially a zero margin of error. FDR laughed all the way to the bank on that one…. The lesson learned here was that the sample had to be truly random and not from a predictable group.
Fast forward to 1948 when there were now national polling companies that promised honest, unbiased results. Almost every national poll smugly pronounced Republican Thomas Dewey to trounce Democrat Harry Truman by a landslide. Every newspaper in the country bit at this tantalizing bait because….well, they’re professional pollsters!! I think we’ve all seen the famous newspaper photo of President Truman holding the Chicago Daily Tribune with the headline “Dewey Defeats Truman”. The lesson: bad sampling, again.
Will they ever learn? My opinion is no. Why would they do that? People have been relying on numbers, stats, and polls for a long time to speak honestly and lend credence to what they were trying to say. Politicians and journalists desperately need these unbiased numbers because no one actually believes what they say. But being the predictable animals that they are, they take these honest, unbiased numbers and then manipulate them to their own end. The result: no one trusts numbers, statistics, or polls anymore. Thanks a lot.
The problem has always been the sample size and the composition of the questions asked. In the 1990’s, it was common that pollsters would only have about a 36% response rate from people called with questions. Care to guess what the pollsters of today use as their average response rate? The Pew Research Center found that only 9% of Americans answered the phone to respond to opinion polls. 9% folks! That’s right, 9% of the population is cited as speaking in the news for all of us. This is the point at which the sample questioned became more important than ever.
And, again, this is just my opinion. But I think that the politicians and journalists actually like this “freedom of interpretation”. They are clearly not looking out for our best interests. They clearly don’t care about us. They want to mold and manipulate us to promote their view. Polls and statistics are simply their method for doing so.
How can we combat this? Disregard them entirely. We voted for them. Now we have to live with them.
We don’t get to discuss this sort of thing in our CHL classes very often. Thanks for indulging me on my blog. If you ever want a CHL class Dallas, TX please let me know. I promise to take care of you