Ok, it’s time to get to the meat and potatoes of this blog. June 2009 arrived and I had only the basic rough draft of my plan to start my new endeavor. I knew it was not going to be anything that took off quickly as my technical skills had to be brought up to speed. It was time to get started, and begin the launching of the website. The would-a, could-a, should- a train, was ready to leave the station.
Several key decisions were made in June. The two largest being, the nature of the online polling company and the other being that I would need to contract the building of the website out to a web design company. A discussion of the actual website and the details of the learning curve to use it will be the topic(s) for future installments of this blog.
It had been apparent to me for several years that most all news, print media, and politicians seemed to confirm their stance on an issue by citing a poll of one type or another. It had been my observation that polls were being conducted by just about anyone; from think tanks, to universities, to news organizations, marketing companies, research companies, and Government organizations. As I researched more , there seemed to be a two- tiered level of acceptance of the accuracy of the polls. The acceptance seems to be based on the whether it can be considered “Scientific” or not. The rest of this installment will detail some of my conclusions and concerns with the way results of polls are conveyed to the public and the elusive description of what is a “Scientific Poll“. “Although I am neither a mathematician or a statistician, there seemed to be some ambiguity as to what scientific approach is acceptable to be considered “Scientific.”
Polls have always intrigued me, whether they were opinion polls or election related polls. I had always gravitated to research that “figured the odds”. When conducting a poll or survey, there is a lot of terminology/jargon used that has become common place, but seems to be so misunderstood. Future Installments to this blog will detail and explore the following terms associated with polls:
Sampling Error
Margin of Error
Selection Bias
Types of Polls (Tracking, Straw, Election, etc.)
Response Bias
Bradley or Wilder effect
Social desirability Bias
More terms will also be covered, but an overview of the abovementioned terms will need to be discussed first. Next Installment 3-7 days.
No comments:
Post a Comment