Poll information can be very deceptive. The very manner in which questions are selected and phrased can profoundly influence the results. That makes them at the very least problematic, and at worst unreliable and therefore utterly useless - unless they are deliberately employed to sway public opinion one way or the other. In that later case they may not be useless (at least not to the proponent) but, in this writer’s opinion, they become even more problematic.

Is it possible, then, to find out what people are really thinking without goading or prodding them with skillfully and often deceptively phrased questions? How do you avoid what we shall call the "questioner's bias" for purposes of this article?

There is indeed a way, and the Internet makes it possible. What if you searched the online keyword-demand for the names of the current presidential candidate roster on both sides of the aisle? The data you'd get back would most certainly be "unscientific" - but that's exactly what is needed. Raw, unfiltered data. The real demand for candidate names as exhibited by actual online searches on the Internet, without goading or prodding. There would be no loaded questions to ask, no way to manipulate anything before it comes back. The only possible kind of manipulation would be the researcher's personal interpretation of the results - after they come back.

So I conducted a little experiment.

Before we begin, though, I have to tell you how I personally filter information. That means I have to tell you how and what I think so you can filter it out as my personal bias. I count myself as an enemy of tyranny in any of its forms - whether individual or collective, socialist or neocon, whether political, social, judicial, or religious. I value freedom of conscience and individual Liberty - if it's combined with individual responsibility. I prefer laissez-faire economics to the endless tinkering and manipulation by so-called "experts" who have no accountability and who couldn't give a whit about the consequences of their (usually) ill-conceived machinations. I believe in limited government a la the founders of this country (minus Alex Hamilton), and I believe war is justified only in defense of ones country - but then should be fought all out, without taking prisoners. You should also know that I am not a Republican and, oh yes, I favor building a reverse-iron curtain to our South, keeping out whoever has no right to be here - according to our criteria, not theirs.

The data below is completely unfiltered, and I make my interpretation of it as obvious as possible. The reason I tell you this is not because I consider it in any way "important" or because I presume that you care. I only tell you what I think so you can effectively filter it out.

The System:

The system I used to gather the reported data is based on Wordtracker.com, an internet search data provider that is most often used by online marketers to ferret out market niches that might be worth exploiting. However, I did not use Wordtracker itself, but another system which uses Wordtracker's data in determining the demand for any given search-keyword or key-phrase (combination of words) used on the Internet in appreciable quantity. That system costs money, which I am sure you don't want to spend just to verify my statements, but you can use Wordtracker's free trial.

You need to understand a bit how Wordtracker works. It does what is called a "vertical search" meaning it detects all search queries (covering over 95% of worldwide search engine use) that were entered during the last month containing the word(s) selected by any particular online information surfers. In the case of "Obama" for example, it returned all search queries used with that word in it, and that in all possible combinations entered. For example, you may get back things like "1 million barack obama president" and "adacity of hope barack obama." This means it brings back even misspelled search terms, as long as they were used by web surfers in appreciable numbers.

It also tells you the cumulative number of how many times each search-phrase was used during the preceding 30 days.

The Methodology:

My initial concern was to find out who among the Republicans was the real front-runner. I had a nagging suspicion that those candidates whom the press presents as having any kind of "electibility" really don't have any - or at least not much of it, in any case. I would then compare the results with the numbers generated by the Democratic front-runners, Barack Obama and Hillary Clinton.

With a few exceptions, I only typed in each candidate's last name to get the largest possible number of returns with all possible permutations. The exceptions are Duncan Hunter, Ron Paul, and John Edwards whose names I entered as first and last name combinations (these surnames are just too common in the world and would return too many unrelated results). Of all results, I picked only the top three or four highest-demand keywords and the related numbers to keep things to a manageable minimum. The numbers stated represent the number of searches Wordtracker measured during the last 30 days.

Here are the results, on a name-search only basis, in descending order:

barack obama 149,790

 1  |  2  |  3  |  4