I used most of my time in this class to analyze political polls in the special New Jersey Senate Election. I am obsessed with politics and enjoy every aspect of them. Polling caught my interest because it was something I didn't understand in the world of politics. I did some background research and I was able to analyze polling charts and backgrounds to help me predict the outcome of the election. Nate Silver's blogs contributed many theories to my project, but I designed my own. I used the New Jersey election as a kind of "guinea pig." After reading a ton about political polling I was able to utilize realclearpolitics.com and print polling charts. I used the sample size, recency, and recognition of the polling agency to rank the polls accordingly. I ranked a pile of charts and went into election day hoping to be correct, and I wasn't. I then got the election results and analyzed why I didn't rank correctly. I came to the conclusion that the locality of a poll also contributes to its correctness. Locality was not an aspect of polling I found in any ranking system out there. This new aspect could effect the entire ranking of polling agencies and their effectiveness for every political race. Locality allows for name recognition of an agency, which could make voters more receptive.
This project was a lot of work, but also a lot of fun. I was glad to find something new to learn in the political world! Polling is just the first step in understanding the political process, I will now continue on and discover new ways to understand momentum and recognition, maybe even political message.
If you are interested in attempting this project here is a link to project instructions: Whose Got the Best Poll?