Poll creation is a very important process. It takes and large amount of time and a long thought process to achieve a properly designed poll. I have learned that question presentation and framing could play a vital role when analyzing the results. It is important to start out with the easy questions in the beginning and work your way to the difficult questions at the end. It may also be beneficial to use screening questions in the survey to find out if your participant is qualified to answer the questions. Doing this will allow for more accurate results.
I have chose to analyze a poll published in the Boomberg News done by Seaton Hall University. My analysis digs further into the concepts of validity, reliability, bias, and agenda of the poll.
This poll was conducted using a sample size of 688 people that were randomly selected, which means that this data can be used to make generalizations to a larger population because it is a probability sample. I think the published poll has a high level of validity. Having selected a random group of 688 people to study provides a large enough number to generalize the data for that entire population. At the end of the article, the source informs the audience that the margin of error is plus or minus 3.8 percent to insure that we know exactly how trusted the data can be. Although the data proves to have a decent level of validity, I do not consider this data to be reliable. The survey was only conducted one time, so we have nothing else to compare the results to. To ensure better reliability, I would need to see a repeated study and calculation of data.
I do not see any bias portray in the published poll. The results were published is a very nonpartisan manner and just reported the results that they have found when analyzing the survey. However, I do not have accesses to the actual questions that were asked in the poll, which may or may not have included leading questions or bias. Based on the information that I was provided, I was not able to detect a bias.
After reading the published poll article, I believe that the sponsor had an agenda of leading the readers to believe that the controversial call in the Packers vs. Seahawks game was the reason to convert people's beliefs of the replacement referees. Using phrases such as "jumped to 65 percent" and quoting the director of the poll saying, "The NFL felt the shift in mood of its public," goes to show what the poll set out to prove.